{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2894906511927788,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 117.859375,
      "epoch": 0.0006447453255963894,
      "grad_norm": 11.706681251525879,
      "kl": 0.0,
      "learning_rate": 9.996776273372017e-07,
      "loss": 0.0,
      "reward": 0.9551776349544525,
      "reward_std": 0.313363179564476,
      "rewards/format_reward": 0.859375,
      "rewards/iou_timestamp_reward": 0.0958026759326458,
      "step": 1
    },
    {
      "completion_length": 119.921875,
      "epoch": 0.0012894906511927789,
      "grad_norm": 9.978516578674316,
      "kl": 0.0011749267578125,
      "learning_rate": 9.993552546744034e-07,
      "loss": 0.0,
      "reward": 0.9701357483863831,
      "reward_std": 0.3779118061065674,
      "rewards/format_reward": 0.875,
      "rewards/iou_timestamp_reward": 0.09513575211167336,
      "step": 2
    },
    {
      "completion_length": 121.859375,
      "epoch": 0.0019342359767891683,
      "grad_norm": 49.87274932861328,
      "kl": 0.001697540283203125,
      "learning_rate": 9.990328820116053e-07,
      "loss": 0.0001,
      "reward": 0.9885792136192322,
      "reward_std": 0.27833086252212524,
      "rewards/format_reward": 0.90625,
      "rewards/iou_timestamp_reward": 0.08232920989394188,
      "step": 3
    },
    {
      "completion_length": 119.546875,
      "epoch": 0.0025789813023855577,
      "grad_norm": 21.538257598876953,
      "kl": 0.003204345703125,
      "learning_rate": 9.987105093488073e-07,
      "loss": 0.0001,
      "reward": 1.1702584028244019,
      "reward_std": 0.3175881803035736,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.21713333204388618,
      "step": 4
    },
    {
      "completion_length": 116.640625,
      "epoch": 0.003223726627981947,
      "grad_norm": 6.138484477996826,
      "kl": 0.0047760009765625,
      "learning_rate": 9.98388136686009e-07,
      "loss": 0.0002,
      "reward": 1.1206515431404114,
      "reward_std": 0.24463944137096405,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.16752655059099197,
      "step": 5
    },
    {
      "completion_length": 117.625,
      "epoch": 0.0038684719535783366,
      "grad_norm": 16.45502471923828,
      "kl": 0.0115966796875,
      "learning_rate": 9.980657640232107e-07,
      "loss": 0.0005,
      "reward": 1.2238141894340515,
      "reward_std": 0.3482338488101959,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.25506413727998734,
      "step": 6
    },
    {
      "completion_length": 108.46875,
      "epoch": 0.004513217279174726,
      "grad_norm": 20.508726119995117,
      "kl": 0.014617919921875,
      "learning_rate": 9.977433913604127e-07,
      "loss": 0.0006,
      "reward": 1.1528931856155396,
      "reward_std": 0.295160673558712,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.19976822286844254,
      "step": 7
    },
    {
      "completion_length": 115.1875,
      "epoch": 0.0051579626047711154,
      "grad_norm": 7.711580276489258,
      "kl": 0.01885986328125,
      "learning_rate": 9.974210186976144e-07,
      "loss": 0.0008,
      "reward": 1.1179298758506775,
      "reward_std": 0.23363575339317322,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.1491798721253872,
      "step": 8
    },
    {
      "completion_length": 117.609375,
      "epoch": 0.005802707930367505,
      "grad_norm": 13.166901588439941,
      "kl": 0.02362060546875,
      "learning_rate": 9.970986460348161e-07,
      "loss": 0.0009,
      "reward": 1.3111140131950378,
      "reward_std": 0.28913620114326477,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.32673898339271545,
      "step": 9
    },
    {
      "completion_length": 116.890625,
      "epoch": 0.006447453255963894,
      "grad_norm": 36.41984176635742,
      "kl": 0.0179443359375,
      "learning_rate": 9.96776273372018e-07,
      "loss": 0.0007,
      "reward": 1.1912025213241577,
      "reward_std": 0.22346250712871552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.19120249524712563,
      "step": 10
    },
    {
      "completion_length": 100.515625,
      "epoch": 0.0070921985815602835,
      "grad_norm": 17.331663131713867,
      "kl": 0.02130126953125,
      "learning_rate": 9.964539007092198e-07,
      "loss": 0.0009,
      "reward": 1.1976637244224548,
      "reward_std": 0.20566103607416153,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.19766373187303543,
      "step": 11
    },
    {
      "completion_length": 108.25,
      "epoch": 0.007736943907156673,
      "grad_norm": 14.938997268676758,
      "kl": 0.0384521484375,
      "learning_rate": 9.961315280464217e-07,
      "loss": 0.0015,
      "reward": 1.0446301102638245,
      "reward_std": 0.2249654456973076,
      "rewards/format_reward": 0.9375,
      "rewards/iou_timestamp_reward": 0.10713006556034088,
      "step": 12
    },
    {
      "completion_length": 109.109375,
      "epoch": 0.008381689232753063,
      "grad_norm": 13.962553977966309,
      "kl": 0.02801513671875,
      "learning_rate": 9.958091553836235e-07,
      "loss": 0.0011,
      "reward": 1.2218610644340515,
      "reward_std": 0.23190322518348694,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.22186104953289032,
      "step": 13
    },
    {
      "completion_length": 112.421875,
      "epoch": 0.009026434558349452,
      "grad_norm": 12.863759994506836,
      "kl": 0.0321044921875,
      "learning_rate": 9.954867827208252e-07,
      "loss": 0.0013,
      "reward": 1.2887030243873596,
      "reward_std": 0.28984855115413666,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.30432797968387604,
      "step": 14
    },
    {
      "completion_length": 94.703125,
      "epoch": 0.009671179883945842,
      "grad_norm": 10.097845077514648,
      "kl": 0.0379638671875,
      "learning_rate": 9.95164410058027e-07,
      "loss": 0.0015,
      "reward": 1.289795994758606,
      "reward_std": 0.23819649964571,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.28979597985744476,
      "step": 15
    },
    {
      "completion_length": 109.71875,
      "epoch": 0.010315925209542231,
      "grad_norm": 14.919957160949707,
      "kl": 0.03289794921875,
      "learning_rate": 9.948420373952289e-07,
      "loss": 0.0013,
      "reward": 1.2771413326263428,
      "reward_std": 0.2302519753575325,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.27714137732982635,
      "step": 16
    },
    {
      "completion_length": 119.546875,
      "epoch": 0.01096067053513862,
      "grad_norm": 7.356471538543701,
      "kl": 0.034912109375,
      "learning_rate": 9.945196647324306e-07,
      "loss": 0.0014,
      "reward": 1.108040452003479,
      "reward_std": 0.19030947610735893,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.1236654780805111,
      "step": 17
    },
    {
      "completion_length": 112.640625,
      "epoch": 0.01160541586073501,
      "grad_norm": 8.942741394042969,
      "kl": 0.03021240234375,
      "learning_rate": 9.941972920696325e-07,
      "loss": 0.0012,
      "reward": 1.2421876192092896,
      "reward_std": 0.25605954229831696,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.27343761920928955,
      "step": 18
    },
    {
      "completion_length": 112.015625,
      "epoch": 0.012250161186331399,
      "grad_norm": 5.671343803405762,
      "kl": 0.037109375,
      "learning_rate": 9.938749194068343e-07,
      "loss": 0.0015,
      "reward": 1.2924596667289734,
      "reward_std": 0.17979642003774643,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2924596965312958,
      "step": 19
    },
    {
      "completion_length": 117.625,
      "epoch": 0.012894906511927788,
      "grad_norm": 7.245625019073486,
      "kl": 0.0355224609375,
      "learning_rate": 9.935525467440362e-07,
      "loss": 0.0014,
      "reward": 1.2731870412826538,
      "reward_std": 0.2262459546327591,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2731870636343956,
      "step": 20
    },
    {
      "completion_length": 110.90625,
      "epoch": 0.013539651837524178,
      "grad_norm": 7.730829238891602,
      "kl": 0.036376953125,
      "learning_rate": 9.93230174081238e-07,
      "loss": 0.0015,
      "reward": 1.2657849192619324,
      "reward_std": 0.25212879478931427,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2657849192619324,
      "step": 21
    },
    {
      "completion_length": 113.953125,
      "epoch": 0.014184397163120567,
      "grad_norm": 8.334545135498047,
      "kl": 0.038330078125,
      "learning_rate": 9.929078014184397e-07,
      "loss": 0.0015,
      "reward": 1.2092431783676147,
      "reward_std": 0.21151791512966156,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.20924309641122818,
      "step": 22
    },
    {
      "completion_length": 100.0625,
      "epoch": 0.014829142488716958,
      "grad_norm": 17.85267448425293,
      "kl": 0.0452880859375,
      "learning_rate": 9.925854287556414e-07,
      "loss": 0.0018,
      "reward": 1.247024118900299,
      "reward_std": 0.2600787580013275,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.24702415615320206,
      "step": 23
    },
    {
      "completion_length": 100.140625,
      "epoch": 0.015473887814313346,
      "grad_norm": 10.987561225891113,
      "kl": 0.048583984375,
      "learning_rate": 9.922630560928433e-07,
      "loss": 0.0019,
      "reward": 1.3182414770126343,
      "reward_std": 0.23921259492635727,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.31824155151844025,
      "step": 24
    },
    {
      "completion_length": 97.75,
      "epoch": 0.016118633139909737,
      "grad_norm": 14.050704956054688,
      "kl": 0.0531005859375,
      "learning_rate": 9.91940683430045e-07,
      "loss": 0.0021,
      "reward": 1.2669792175292969,
      "reward_std": 0.22845448553562164,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2669791951775551,
      "step": 25
    },
    {
      "completion_length": 101.078125,
      "epoch": 0.016763378465506126,
      "grad_norm": 12.309895515441895,
      "kl": 0.0396728515625,
      "learning_rate": 9.91618310767247e-07,
      "loss": 0.0016,
      "reward": 1.2797068357467651,
      "reward_std": 0.1980849951505661,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.27970677614212036,
      "step": 26
    },
    {
      "completion_length": 84.90625,
      "epoch": 0.017408123791102514,
      "grad_norm": 8.08408260345459,
      "kl": 0.0487060546875,
      "learning_rate": 9.912959381044487e-07,
      "loss": 0.0019,
      "reward": 1.2797399759292603,
      "reward_std": 0.1894526183605194,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.29536499083042145,
      "step": 27
    },
    {
      "completion_length": 100.75,
      "epoch": 0.018052869116698903,
      "grad_norm": 13.173178672790527,
      "kl": 0.048095703125,
      "learning_rate": 9.909735654416504e-07,
      "loss": 0.0019,
      "reward": 1.254739224910736,
      "reward_std": 0.19636224955320358,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.2703641951084137,
      "step": 28
    },
    {
      "completion_length": 90.0625,
      "epoch": 0.018697614442295292,
      "grad_norm": 14.946029663085938,
      "kl": 0.0457763671875,
      "learning_rate": 9.906511927788524e-07,
      "loss": 0.0018,
      "reward": 1.2612407207489014,
      "reward_std": 0.19902688264846802,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2612406611442566,
      "step": 29
    },
    {
      "completion_length": 92.875,
      "epoch": 0.019342359767891684,
      "grad_norm": 24.124683380126953,
      "kl": 0.04345703125,
      "learning_rate": 9.903288201160541e-07,
      "loss": 0.0017,
      "reward": 1.3272080421447754,
      "reward_std": 0.24419597536325455,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.32720808684825897,
      "step": 30
    },
    {
      "completion_length": 92.421875,
      "epoch": 0.019987105093488073,
      "grad_norm": 19.198226928710938,
      "kl": 0.1097412109375,
      "learning_rate": 9.900064474532558e-07,
      "loss": 0.0044,
      "reward": 1.3916027545928955,
      "reward_std": 0.2533900588750839,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4072277396917343,
      "step": 31
    },
    {
      "completion_length": 101.421875,
      "epoch": 0.020631850419084462,
      "grad_norm": 9.14734172821045,
      "kl": 0.0443115234375,
      "learning_rate": 9.896840747904578e-07,
      "loss": 0.0018,
      "reward": 1.247195303440094,
      "reward_std": 0.2349850833415985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2471952885389328,
      "step": 32
    },
    {
      "completion_length": 95.921875,
      "epoch": 0.02127659574468085,
      "grad_norm": 8.344867706298828,
      "kl": 0.037841796875,
      "learning_rate": 9.893617021276595e-07,
      "loss": 0.0015,
      "reward": 1.343221127986908,
      "reward_std": 0.20855669677257538,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.34322114288806915,
      "step": 33
    },
    {
      "completion_length": 93.359375,
      "epoch": 0.02192134107027724,
      "grad_norm": 19.839126586914062,
      "kl": 0.0396728515625,
      "learning_rate": 9.890393294648614e-07,
      "loss": 0.0016,
      "reward": 1.2119425535202026,
      "reward_std": 0.1833260953426361,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.21194250136613846,
      "step": 34
    },
    {
      "completion_length": 112.921875,
      "epoch": 0.02256608639587363,
      "grad_norm": 10.852544784545898,
      "kl": 0.0394287109375,
      "learning_rate": 9.887169568020632e-07,
      "loss": 0.0016,
      "reward": 1.2822151184082031,
      "reward_std": 0.19346018135547638,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.28221510350704193,
      "step": 35
    },
    {
      "completion_length": 110.859375,
      "epoch": 0.02321083172147002,
      "grad_norm": 66.1825942993164,
      "kl": 0.03253173828125,
      "learning_rate": 9.88394584139265e-07,
      "loss": 0.0013,
      "reward": 1.2645833492279053,
      "reward_std": 0.22069542109966278,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.26458336412906647,
      "step": 36
    },
    {
      "completion_length": 104.421875,
      "epoch": 0.02385557704706641,
      "grad_norm": 12.111766815185547,
      "kl": 0.0426025390625,
      "learning_rate": 9.880722114764668e-07,
      "loss": 0.0017,
      "reward": 1.2878879308700562,
      "reward_std": 0.21032921224832535,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.28788794577121735,
      "step": 37
    },
    {
      "completion_length": 110.65625,
      "epoch": 0.024500322372662798,
      "grad_norm": 11.206411361694336,
      "kl": 0.0400390625,
      "learning_rate": 9.877498388136686e-07,
      "loss": 0.0016,
      "reward": 1.3537410497665405,
      "reward_std": 0.2186509370803833,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3537410721182823,
      "step": 38
    },
    {
      "completion_length": 111.984375,
      "epoch": 0.025145067698259187,
      "grad_norm": 17.833251953125,
      "kl": 0.0408935546875,
      "learning_rate": 9.874274661508703e-07,
      "loss": 0.0016,
      "reward": 1.3208078145980835,
      "reward_std": 0.21878959983587265,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3208078145980835,
      "step": 39
    },
    {
      "completion_length": 119.75,
      "epoch": 0.025789813023855575,
      "grad_norm": 40.49732208251953,
      "kl": 0.334716796875,
      "learning_rate": 9.871050934880722e-07,
      "loss": 0.0134,
      "reward": 1.23319411277771,
      "reward_std": 0.2120913863182068,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.23319409787654877,
      "step": 40
    },
    {
      "completion_length": 104.25,
      "epoch": 0.026434558349451968,
      "grad_norm": 19.549007415771484,
      "kl": 0.050537109375,
      "learning_rate": 9.86782720825274e-07,
      "loss": 0.002,
      "reward": 1.307875633239746,
      "reward_std": 0.16077548265457153,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3078755736351013,
      "step": 41
    },
    {
      "completion_length": 105.921875,
      "epoch": 0.027079303675048357,
      "grad_norm": 29.055553436279297,
      "kl": 0.056640625,
      "learning_rate": 9.86460348162476e-07,
      "loss": 0.0023,
      "reward": 1.3375996351242065,
      "reward_std": 0.23000206798315048,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33759957551956177,
      "step": 42
    },
    {
      "completion_length": 103.140625,
      "epoch": 0.027724049000644745,
      "grad_norm": 7.140056610107422,
      "kl": 0.101318359375,
      "learning_rate": 9.861379754996776e-07,
      "loss": 0.0041,
      "reward": 1.2600332498550415,
      "reward_std": 0.2108185514807701,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2600333169102669,
      "step": 43
    },
    {
      "completion_length": 100.625,
      "epoch": 0.028368794326241134,
      "grad_norm": 36.15247344970703,
      "kl": 0.06201171875,
      "learning_rate": 9.858156028368794e-07,
      "loss": 0.0025,
      "reward": 1.2336326837539673,
      "reward_std": 0.17267921566963196,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2336326688528061,
      "step": 44
    },
    {
      "completion_length": 98.484375,
      "epoch": 0.029013539651837523,
      "grad_norm": 12.655134201049805,
      "kl": 0.052490234375,
      "learning_rate": 9.85493230174081e-07,
      "loss": 0.0021,
      "reward": 1.4318326711654663,
      "reward_std": 0.20957604050636292,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4318326860666275,
      "step": 45
    },
    {
      "completion_length": 102.71875,
      "epoch": 0.029658284977433915,
      "grad_norm": 25.280502319335938,
      "kl": 0.059326171875,
      "learning_rate": 9.85170857511283e-07,
      "loss": 0.0024,
      "reward": 1.3532536029815674,
      "reward_std": 0.21924903988838196,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3532535433769226,
      "step": 46
    },
    {
      "completion_length": 104.140625,
      "epoch": 0.030303030303030304,
      "grad_norm": 10.246079444885254,
      "kl": 0.0511474609375,
      "learning_rate": 9.848484848484847e-07,
      "loss": 0.002,
      "reward": 1.4125980138778687,
      "reward_std": 0.1969481185078621,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41259796917438507,
      "step": 47
    },
    {
      "completion_length": 96.484375,
      "epoch": 0.030947775628626693,
      "grad_norm": 32.44934844970703,
      "kl": 0.0533447265625,
      "learning_rate": 9.845261121856867e-07,
      "loss": 0.0021,
      "reward": 1.3488176465034485,
      "reward_std": 0.2110263928771019,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.3644426167011261,
      "step": 48
    },
    {
      "completion_length": 94.328125,
      "epoch": 0.03159252095422308,
      "grad_norm": 6.3295698165893555,
      "kl": 0.0740966796875,
      "learning_rate": 9.842037395228884e-07,
      "loss": 0.003,
      "reward": 1.3096264600753784,
      "reward_std": 0.1813429892063141,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3096264898777008,
      "step": 49
    },
    {
      "completion_length": 96.25,
      "epoch": 0.032237266279819474,
      "grad_norm": 12.89992618560791,
      "kl": 0.0531005859375,
      "learning_rate": 9.838813668600904e-07,
      "loss": 0.0021,
      "reward": 1.3683323860168457,
      "reward_std": 0.2907109409570694,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.39958229660987854,
      "step": 50
    },
    {
      "completion_length": 90.84375,
      "epoch": 0.03288201160541586,
      "grad_norm": 106.75090789794922,
      "kl": 0.0523681640625,
      "learning_rate": 9.83558994197292e-07,
      "loss": 0.0021,
      "reward": 1.4628892540931702,
      "reward_std": 0.22101883590221405,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46288928389549255,
      "step": 51
    },
    {
      "completion_length": 84.015625,
      "epoch": 0.03352675693101225,
      "grad_norm": 8.15224552154541,
      "kl": 0.060302734375,
      "learning_rate": 9.832366215344938e-07,
      "loss": 0.0024,
      "reward": 1.36749267578125,
      "reward_std": 0.22733061015605927,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3674926608800888,
      "step": 52
    },
    {
      "completion_length": 88.3125,
      "epoch": 0.03417150225660864,
      "grad_norm": 10.359516143798828,
      "kl": 0.055908203125,
      "learning_rate": 9.829142488716955e-07,
      "loss": 0.0022,
      "reward": 1.3939309120178223,
      "reward_std": 0.18551449477672577,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39393097162246704,
      "step": 53
    },
    {
      "completion_length": 82.8125,
      "epoch": 0.03481624758220503,
      "grad_norm": 23.30002212524414,
      "kl": 0.0635986328125,
      "learning_rate": 9.825918762088975e-07,
      "loss": 0.0025,
      "reward": 1.3875991702079773,
      "reward_std": 0.2218652218580246,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3875991553068161,
      "step": 54
    },
    {
      "completion_length": 84.015625,
      "epoch": 0.03546099290780142,
      "grad_norm": 9.210583686828613,
      "kl": 0.0657958984375,
      "learning_rate": 9.822695035460992e-07,
      "loss": 0.0026,
      "reward": 1.3397376537322998,
      "reward_std": 0.2256728559732437,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.3709876760840416,
      "step": 55
    },
    {
      "completion_length": 78.359375,
      "epoch": 0.036105738233397806,
      "grad_norm": 18.673093795776367,
      "kl": 0.074951171875,
      "learning_rate": 9.819471308833011e-07,
      "loss": 0.003,
      "reward": 1.3963087797164917,
      "reward_std": 0.2275141477584839,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3963087424635887,
      "step": 56
    },
    {
      "completion_length": 86.234375,
      "epoch": 0.0367504835589942,
      "grad_norm": 21.288368225097656,
      "kl": 0.091552734375,
      "learning_rate": 9.816247582205029e-07,
      "loss": 0.0037,
      "reward": 1.4477627873420715,
      "reward_std": 0.194952093064785,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44776275753974915,
      "step": 57
    },
    {
      "completion_length": 82.671875,
      "epoch": 0.037395228884590584,
      "grad_norm": 12.072967529296875,
      "kl": 0.079345703125,
      "learning_rate": 9.813023855577046e-07,
      "loss": 0.0032,
      "reward": 1.5084742307662964,
      "reward_std": 0.15730255097150803,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5084741413593292,
      "step": 58
    },
    {
      "completion_length": 76.5625,
      "epoch": 0.038039974210186976,
      "grad_norm": 12.807083129882812,
      "kl": 0.067626953125,
      "learning_rate": 9.809800128949065e-07,
      "loss": 0.0027,
      "reward": 1.396537721157074,
      "reward_std": 0.21208080649375916,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3965377062559128,
      "step": 59
    },
    {
      "completion_length": 78.0,
      "epoch": 0.03868471953578337,
      "grad_norm": 32.63819885253906,
      "kl": 0.074951171875,
      "learning_rate": 9.806576402321083e-07,
      "loss": 0.003,
      "reward": 1.4429082870483398,
      "reward_std": 0.19335371255874634,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44290828704833984,
      "step": 60
    },
    {
      "completion_length": 81.71875,
      "epoch": 0.039329464861379754,
      "grad_norm": 7.923285961151123,
      "kl": 0.0810546875,
      "learning_rate": 9.8033526756931e-07,
      "loss": 0.0032,
      "reward": 1.2859596014022827,
      "reward_std": 0.19862326979637146,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2859596461057663,
      "step": 61
    },
    {
      "completion_length": 73.09375,
      "epoch": 0.039974210186976146,
      "grad_norm": 45.514076232910156,
      "kl": 0.0791015625,
      "learning_rate": 9.80012894906512e-07,
      "loss": 0.0032,
      "reward": 1.3525552153587341,
      "reward_std": 0.1816207319498062,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3525552302598953,
      "step": 62
    },
    {
      "completion_length": 73.234375,
      "epoch": 0.04061895551257253,
      "grad_norm": 25.02358627319336,
      "kl": 0.0810546875,
      "learning_rate": 9.796905222437137e-07,
      "loss": 0.0032,
      "reward": 1.3032844066619873,
      "reward_std": 0.2011517509818077,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3032844811677933,
      "step": 63
    },
    {
      "completion_length": 77.375,
      "epoch": 0.041263700838168924,
      "grad_norm": 18.431140899658203,
      "kl": 0.092041015625,
      "learning_rate": 9.793681495809156e-07,
      "loss": 0.0037,
      "reward": 1.4467111825942993,
      "reward_std": 0.23784177750349045,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44671112298965454,
      "step": 64
    },
    {
      "completion_length": 76.78125,
      "epoch": 0.041908446163765316,
      "grad_norm": 10.069541931152344,
      "kl": 0.0888671875,
      "learning_rate": 9.790457769181173e-07,
      "loss": 0.0036,
      "reward": 1.4564274549484253,
      "reward_std": 0.24415691941976547,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4564273804426193,
      "step": 65
    },
    {
      "completion_length": 79.359375,
      "epoch": 0.0425531914893617,
      "grad_norm": 8.669953346252441,
      "kl": 0.09521484375,
      "learning_rate": 9.78723404255319e-07,
      "loss": 0.0038,
      "reward": 1.3141262531280518,
      "reward_std": 0.21810634434223175,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.32975125312805176,
      "step": 66
    },
    {
      "completion_length": 79.953125,
      "epoch": 0.04319793681495809,
      "grad_norm": 22.783241271972656,
      "kl": 0.080078125,
      "learning_rate": 9.78401031592521e-07,
      "loss": 0.0032,
      "reward": 1.4575133919715881,
      "reward_std": 0.18917211890220642,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4575134366750717,
      "step": 67
    },
    {
      "completion_length": 78.296875,
      "epoch": 0.04384268214055448,
      "grad_norm": 22.941936492919922,
      "kl": 0.09326171875,
      "learning_rate": 9.780786589297227e-07,
      "loss": 0.0037,
      "reward": 1.4472717642784119,
      "reward_std": 0.15321099758148193,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4472717046737671,
      "step": 68
    },
    {
      "completion_length": 78.640625,
      "epoch": 0.04448742746615087,
      "grad_norm": 17.53478240966797,
      "kl": 0.091552734375,
      "learning_rate": 9.777562862669244e-07,
      "loss": 0.0037,
      "reward": 1.265991985797882,
      "reward_std": 0.18441501259803772,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.28161704540252686,
      "step": 69
    },
    {
      "completion_length": 71.9375,
      "epoch": 0.04513217279174726,
      "grad_norm": 12.067954063415527,
      "kl": 0.1015625,
      "learning_rate": 9.774339136041264e-07,
      "loss": 0.0041,
      "reward": 1.4680742025375366,
      "reward_std": 0.17444680631160736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4680742621421814,
      "step": 70
    },
    {
      "completion_length": 73.640625,
      "epoch": 0.04577691811734365,
      "grad_norm": 12.765181541442871,
      "kl": 0.115478515625,
      "learning_rate": 9.771115409413281e-07,
      "loss": 0.0046,
      "reward": 1.384080708026886,
      "reward_std": 0.2056313455104828,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.39970576763153076,
      "step": 71
    },
    {
      "completion_length": 74.9375,
      "epoch": 0.04642166344294004,
      "grad_norm": 15.447208404541016,
      "kl": 0.088623046875,
      "learning_rate": 9.7678916827853e-07,
      "loss": 0.0035,
      "reward": 1.3486806750297546,
      "reward_std": 0.2008630558848381,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.34868063032627106,
      "step": 72
    },
    {
      "completion_length": 72.875,
      "epoch": 0.047066408768536426,
      "grad_norm": 12.092300415039062,
      "kl": 0.1162109375,
      "learning_rate": 9.764667956157318e-07,
      "loss": 0.0047,
      "reward": 1.413942575454712,
      "reward_std": 0.2041231021285057,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4139425456523895,
      "step": 73
    },
    {
      "completion_length": 89.46875,
      "epoch": 0.04771115409413282,
      "grad_norm": 16.925933837890625,
      "kl": 0.070068359375,
      "learning_rate": 9.761444229529335e-07,
      "loss": 0.0028,
      "reward": 1.3617019653320312,
      "reward_std": 0.16857799887657166,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3617018908262253,
      "step": 74
    },
    {
      "completion_length": 80.109375,
      "epoch": 0.048355899419729204,
      "grad_norm": 18.16619110107422,
      "kl": 0.0762939453125,
      "learning_rate": 9.758220502901352e-07,
      "loss": 0.0031,
      "reward": 1.3572936058044434,
      "reward_std": 0.22740349918603897,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.37291859090328217,
      "step": 75
    },
    {
      "completion_length": 77.640625,
      "epoch": 0.049000644745325596,
      "grad_norm": 40.9599609375,
      "kl": 0.0771484375,
      "learning_rate": 9.754996776273372e-07,
      "loss": 0.0031,
      "reward": 1.3504416346549988,
      "reward_std": 0.18997812271118164,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35044166445732117,
      "step": 76
    },
    {
      "completion_length": 77.640625,
      "epoch": 0.04964539007092199,
      "grad_norm": 9.37459945678711,
      "kl": 0.09033203125,
      "learning_rate": 9.75177304964539e-07,
      "loss": 0.0036,
      "reward": 1.375741422176361,
      "reward_std": 0.20171231776475906,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3757414370775223,
      "step": 77
    },
    {
      "completion_length": 83.421875,
      "epoch": 0.05029013539651837,
      "grad_norm": 8.924232482910156,
      "kl": 0.07861328125,
      "learning_rate": 9.748549323017408e-07,
      "loss": 0.0031,
      "reward": 1.3597639203071594,
      "reward_std": 0.21023260056972504,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.3753889948129654,
      "step": 78
    },
    {
      "completion_length": 83.703125,
      "epoch": 0.050934880722114766,
      "grad_norm": 10.10476016998291,
      "kl": 0.077880859375,
      "learning_rate": 9.745325596389426e-07,
      "loss": 0.0031,
      "reward": 1.273858666419983,
      "reward_std": 0.18880721181631088,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2738586515188217,
      "step": 79
    },
    {
      "completion_length": 79.109375,
      "epoch": 0.05157962604771115,
      "grad_norm": 11.688029289245605,
      "kl": 0.083251953125,
      "learning_rate": 9.742101869761445e-07,
      "loss": 0.0033,
      "reward": 1.4125596284866333,
      "reward_std": 0.16709274798631668,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4125596284866333,
      "step": 80
    },
    {
      "completion_length": 75.21875,
      "epoch": 0.05222437137330754,
      "grad_norm": 14.252153396606445,
      "kl": 0.085205078125,
      "learning_rate": 9.738878143133462e-07,
      "loss": 0.0034,
      "reward": 1.355860710144043,
      "reward_std": 0.25486576557159424,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3558606505393982,
      "step": 81
    },
    {
      "completion_length": 82.0625,
      "epoch": 0.052869116698903935,
      "grad_norm": 23.381254196166992,
      "kl": 0.089111328125,
      "learning_rate": 9.73565441650548e-07,
      "loss": 0.0036,
      "reward": 1.4577891826629639,
      "reward_std": 0.222563698887825,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45778918266296387,
      "step": 82
    },
    {
      "completion_length": 85.671875,
      "epoch": 0.05351386202450032,
      "grad_norm": 29.149097442626953,
      "kl": 0.076904296875,
      "learning_rate": 9.732430689877497e-07,
      "loss": 0.0031,
      "reward": 1.3724395632743835,
      "reward_std": 0.16367345303297043,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37243956327438354,
      "step": 83
    },
    {
      "completion_length": 86.78125,
      "epoch": 0.05415860735009671,
      "grad_norm": 11.676203727722168,
      "kl": 0.083984375,
      "learning_rate": 9.729206963249516e-07,
      "loss": 0.0034,
      "reward": 1.4838296175003052,
      "reward_std": 0.2798515185713768,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4994546175003052,
      "step": 84
    },
    {
      "completion_length": 77.171875,
      "epoch": 0.0548033526756931,
      "grad_norm": 9.494368553161621,
      "kl": 0.0966796875,
      "learning_rate": 9.725983236621534e-07,
      "loss": 0.0039,
      "reward": 1.3889164328575134,
      "reward_std": 0.18041953444480896,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.38891640305519104,
      "step": 85
    },
    {
      "completion_length": 79.171875,
      "epoch": 0.05544809800128949,
      "grad_norm": 15.85273265838623,
      "kl": 0.082763671875,
      "learning_rate": 9.722759509993553e-07,
      "loss": 0.0033,
      "reward": 1.2888696193695068,
      "reward_std": 0.18666045367717743,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2888696789741516,
      "step": 86
    },
    {
      "completion_length": 80.375,
      "epoch": 0.05609284332688588,
      "grad_norm": 33.20154571533203,
      "kl": 0.09130859375,
      "learning_rate": 9.71953578336557e-07,
      "loss": 0.0037,
      "reward": 1.4762005805969238,
      "reward_std": 0.20586346834897995,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.476200670003891,
      "step": 87
    },
    {
      "completion_length": 80.25,
      "epoch": 0.05673758865248227,
      "grad_norm": 23.280982971191406,
      "kl": 0.0849609375,
      "learning_rate": 9.716312056737588e-07,
      "loss": 0.0034,
      "reward": 1.3444839119911194,
      "reward_std": 0.18448061496019363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3444839417934418,
      "step": 88
    },
    {
      "completion_length": 86.203125,
      "epoch": 0.05738233397807866,
      "grad_norm": 7.881406307220459,
      "kl": 0.07373046875,
      "learning_rate": 9.713088330109607e-07,
      "loss": 0.003,
      "reward": 1.4067343473434448,
      "reward_std": 0.18169286847114563,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.406734362244606,
      "step": 89
    },
    {
      "completion_length": 81.0625,
      "epoch": 0.058027079303675046,
      "grad_norm": 13.038424491882324,
      "kl": 0.084228515625,
      "learning_rate": 9.709864603481624e-07,
      "loss": 0.0034,
      "reward": 1.4451138973236084,
      "reward_std": 0.21713434159755707,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4451138973236084,
      "step": 90
    },
    {
      "completion_length": 79.609375,
      "epoch": 0.05867182462927144,
      "grad_norm": 7.999779224395752,
      "kl": 0.081298828125,
      "learning_rate": 9.706640876853641e-07,
      "loss": 0.0033,
      "reward": 1.40691077709198,
      "reward_std": 0.2112068235874176,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4069107621908188,
      "step": 91
    },
    {
      "completion_length": 73.28125,
      "epoch": 0.05931656995486783,
      "grad_norm": 19.086753845214844,
      "kl": 0.11572265625,
      "learning_rate": 9.70341715022566e-07,
      "loss": 0.0046,
      "reward": 1.4588000178337097,
      "reward_std": 0.1944720521569252,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4588000625371933,
      "step": 92
    },
    {
      "completion_length": 83.25,
      "epoch": 0.059961315280464215,
      "grad_norm": 24.621259689331055,
      "kl": 0.101806640625,
      "learning_rate": 9.700193423597678e-07,
      "loss": 0.0041,
      "reward": 1.4452659487724304,
      "reward_std": 0.21456287056207657,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4452659338712692,
      "step": 93
    },
    {
      "completion_length": 76.28125,
      "epoch": 0.06060606060606061,
      "grad_norm": 18.733131408691406,
      "kl": 0.088623046875,
      "learning_rate": 9.696969696969698e-07,
      "loss": 0.0035,
      "reward": 1.4556202292442322,
      "reward_std": 0.11773604899644852,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.455620214343071,
      "step": 94
    },
    {
      "completion_length": 73.015625,
      "epoch": 0.06125080593165699,
      "grad_norm": 11.115130424499512,
      "kl": 0.096435546875,
      "learning_rate": 9.693745970341715e-07,
      "loss": 0.0039,
      "reward": 1.3447935581207275,
      "reward_std": 0.18323758244514465,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.34479353576898575,
      "step": 95
    },
    {
      "completion_length": 73.3125,
      "epoch": 0.061895551257253385,
      "grad_norm": 15.140289306640625,
      "kl": 0.093994140625,
      "learning_rate": 9.690522243713732e-07,
      "loss": 0.0038,
      "reward": 1.475332260131836,
      "reward_std": 0.23074129223823547,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4753323048353195,
      "step": 96
    },
    {
      "completion_length": 74.53125,
      "epoch": 0.06254029658284978,
      "grad_norm": 23.077402114868164,
      "kl": 0.10498046875,
      "learning_rate": 9.68729851708575e-07,
      "loss": 0.0042,
      "reward": 1.417439877986908,
      "reward_std": 0.17361952364444733,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41743984818458557,
      "step": 97
    },
    {
      "completion_length": 73.5,
      "epoch": 0.06318504190844616,
      "grad_norm": 13.471073150634766,
      "kl": 0.099853515625,
      "learning_rate": 9.684074790457769e-07,
      "loss": 0.004,
      "reward": 1.3859241008758545,
      "reward_std": 0.20852358639240265,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3859241306781769,
      "step": 98
    },
    {
      "completion_length": 72.25,
      "epoch": 0.06382978723404255,
      "grad_norm": 12.648615837097168,
      "kl": 0.093017578125,
      "learning_rate": 9.680851063829786e-07,
      "loss": 0.0037,
      "reward": 1.3480225801467896,
      "reward_std": 0.22543464601039886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.34802258014678955,
      "step": 99
    },
    {
      "completion_length": 71.875,
      "epoch": 0.06447453255963895,
      "grad_norm": 12.842320442199707,
      "kl": 0.080810546875,
      "learning_rate": 9.677627337201805e-07,
      "loss": 0.0032,
      "reward": 1.4175334572792053,
      "reward_std": 0.18312367051839828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4175334572792053,
      "step": 100
    },
    {
      "completion_length": 70.515625,
      "epoch": 0.06511927788523533,
      "grad_norm": 13.96192455291748,
      "kl": 0.106201171875,
      "learning_rate": 9.674403610573823e-07,
      "loss": 0.0042,
      "reward": 1.3348948955535889,
      "reward_std": 0.166929941624403,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3348948061466217,
      "step": 101
    },
    {
      "completion_length": 71.375,
      "epoch": 0.06576402321083172,
      "grad_norm": 12.400444984436035,
      "kl": 0.12841796875,
      "learning_rate": 9.671179883945842e-07,
      "loss": 0.0051,
      "reward": 1.4501676559448242,
      "reward_std": 0.180241659283638,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45016759634017944,
      "step": 102
    },
    {
      "completion_length": 71.3125,
      "epoch": 0.06640876853642812,
      "grad_norm": 23.09809112548828,
      "kl": 0.1064453125,
      "learning_rate": 9.66795615731786e-07,
      "loss": 0.0043,
      "reward": 1.4367554187774658,
      "reward_std": 0.12167369574308395,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43675535917282104,
      "step": 103
    },
    {
      "completion_length": 72.09375,
      "epoch": 0.0670535138620245,
      "grad_norm": 7.730201244354248,
      "kl": 0.104736328125,
      "learning_rate": 9.664732430689877e-07,
      "loss": 0.0042,
      "reward": 1.4592158794403076,
      "reward_std": 0.193368099629879,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4592158794403076,
      "step": 104
    },
    {
      "completion_length": 68.953125,
      "epoch": 0.06769825918762089,
      "grad_norm": 17.011564254760742,
      "kl": 0.1201171875,
      "learning_rate": 9.661508704061894e-07,
      "loss": 0.0048,
      "reward": 1.3875025510787964,
      "reward_std": 0.1890907734632492,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.38750262558460236,
      "step": 105
    },
    {
      "completion_length": 76.875,
      "epoch": 0.06834300451321727,
      "grad_norm": 20.02046012878418,
      "kl": 0.0947265625,
      "learning_rate": 9.658284977433913e-07,
      "loss": 0.0038,
      "reward": 1.5030284523963928,
      "reward_std": 0.16404563933610916,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5030284374952316,
      "step": 106
    },
    {
      "completion_length": 75.0,
      "epoch": 0.06898774983881367,
      "grad_norm": 13.054033279418945,
      "kl": 0.1015625,
      "learning_rate": 9.65506125080593e-07,
      "loss": 0.0041,
      "reward": 1.571265459060669,
      "reward_std": 0.16624636203050613,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5712654292583466,
      "step": 107
    },
    {
      "completion_length": 70.609375,
      "epoch": 0.06963249516441006,
      "grad_norm": 30.342504501342773,
      "kl": 0.11669921875,
      "learning_rate": 9.65183752417795e-07,
      "loss": 0.0047,
      "reward": 1.4218077063560486,
      "reward_std": 0.20463600754737854,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4218077063560486,
      "step": 108
    },
    {
      "completion_length": 75.953125,
      "epoch": 0.07027724049000644,
      "grad_norm": 14.21203899383545,
      "kl": 0.09912109375,
      "learning_rate": 9.648613797549967e-07,
      "loss": 0.004,
      "reward": 1.4889991879463196,
      "reward_std": 0.18232028931379318,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4889991283416748,
      "step": 109
    },
    {
      "completion_length": 77.125,
      "epoch": 0.07092198581560284,
      "grad_norm": 16.088598251342773,
      "kl": 0.1142578125,
      "learning_rate": 9.645390070921985e-07,
      "loss": 0.0046,
      "reward": 1.4165233969688416,
      "reward_std": 0.18773652613162994,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41652336716651917,
      "step": 110
    },
    {
      "completion_length": 73.53125,
      "epoch": 0.07156673114119923,
      "grad_norm": 13.502654075622559,
      "kl": 0.10009765625,
      "learning_rate": 9.642166344294004e-07,
      "loss": 0.004,
      "reward": 1.4663327932357788,
      "reward_std": 0.17598994821310043,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4663327932357788,
      "step": 111
    },
    {
      "completion_length": 70.34375,
      "epoch": 0.07221147646679561,
      "grad_norm": 14.628576278686523,
      "kl": 0.102294921875,
      "learning_rate": 9.638942617666021e-07,
      "loss": 0.0041,
      "reward": 1.4488741755485535,
      "reward_std": 0.17282917350530624,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4488741457462311,
      "step": 112
    },
    {
      "completion_length": 75.375,
      "epoch": 0.07285622179239201,
      "grad_norm": 10.183534622192383,
      "kl": 0.103271484375,
      "learning_rate": 9.635718891038038e-07,
      "loss": 0.0041,
      "reward": 1.390573263168335,
      "reward_std": 0.1438509002327919,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39057329297065735,
      "step": 113
    },
    {
      "completion_length": 72.375,
      "epoch": 0.0735009671179884,
      "grad_norm": 15.171917915344238,
      "kl": 0.130615234375,
      "learning_rate": 9.632495164410058e-07,
      "loss": 0.0052,
      "reward": 1.4449996948242188,
      "reward_std": 0.15270018205046654,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4449998140335083,
      "step": 114
    },
    {
      "completion_length": 67.640625,
      "epoch": 0.07414571244358478,
      "grad_norm": 15.202000617980957,
      "kl": 0.14111328125,
      "learning_rate": 9.629271437782075e-07,
      "loss": 0.0056,
      "reward": 1.4991153478622437,
      "reward_std": 0.1308480128645897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49911533296108246,
      "step": 115
    },
    {
      "completion_length": 72.390625,
      "epoch": 0.07479045776918117,
      "grad_norm": 11.967129707336426,
      "kl": 0.1357421875,
      "learning_rate": 9.626047711154095e-07,
      "loss": 0.0054,
      "reward": 1.4869657158851624,
      "reward_std": 0.1620316430926323,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4869658052921295,
      "step": 116
    },
    {
      "completion_length": 76.890625,
      "epoch": 0.07543520309477757,
      "grad_norm": 15.96191692352295,
      "kl": 0.1484375,
      "learning_rate": 9.622823984526112e-07,
      "loss": 0.0059,
      "reward": 1.5816780924797058,
      "reward_std": 0.18133046478033066,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.581678032875061,
      "step": 117
    },
    {
      "completion_length": 75.53125,
      "epoch": 0.07607994842037395,
      "grad_norm": 12.706347465515137,
      "kl": 0.11328125,
      "learning_rate": 9.61960025789813e-07,
      "loss": 0.0045,
      "reward": 1.5155820846557617,
      "reward_std": 0.13779311254620552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5155820995569229,
      "step": 118
    },
    {
      "completion_length": 75.921875,
      "epoch": 0.07672469374597034,
      "grad_norm": 10.711260795593262,
      "kl": 0.12451171875,
      "learning_rate": 9.616376531270149e-07,
      "loss": 0.005,
      "reward": 1.4439030289649963,
      "reward_std": 0.1384158730506897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44390299916267395,
      "step": 119
    },
    {
      "completion_length": 77.921875,
      "epoch": 0.07736943907156674,
      "grad_norm": 8.561759948730469,
      "kl": 0.118896484375,
      "learning_rate": 9.613152804642166e-07,
      "loss": 0.0048,
      "reward": 1.6015077829360962,
      "reward_std": 0.187755785882473,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.601507842540741,
      "step": 120
    },
    {
      "completion_length": 76.328125,
      "epoch": 0.07801418439716312,
      "grad_norm": 21.100255966186523,
      "kl": 0.4189453125,
      "learning_rate": 9.609929078014183e-07,
      "loss": 0.0168,
      "reward": 1.5190616846084595,
      "reward_std": 0.1306452825665474,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5190616846084595,
      "step": 121
    },
    {
      "completion_length": 75.78125,
      "epoch": 0.07865892972275951,
      "grad_norm": 32.30814743041992,
      "kl": 0.14404296875,
      "learning_rate": 9.606705351386202e-07,
      "loss": 0.0058,
      "reward": 1.394041121006012,
      "reward_std": 0.16053549945354462,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39404113590717316,
      "step": 122
    },
    {
      "completion_length": 71.28125,
      "epoch": 0.07930367504835589,
      "grad_norm": 18.817506790161133,
      "kl": 0.15673828125,
      "learning_rate": 9.60348162475822e-07,
      "loss": 0.0063,
      "reward": 1.4487716555595398,
      "reward_std": 0.24345475435256958,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.464396670460701,
      "step": 123
    },
    {
      "completion_length": 69.875,
      "epoch": 0.07994842037395229,
      "grad_norm": 16.40485382080078,
      "kl": 0.14306640625,
      "learning_rate": 9.60025789813024e-07,
      "loss": 0.0057,
      "reward": 1.5178059339523315,
      "reward_std": 0.14177018404006958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5178059637546539,
      "step": 124
    },
    {
      "completion_length": 72.15625,
      "epoch": 0.08059316569954868,
      "grad_norm": 9.994036674499512,
      "kl": 0.1435546875,
      "learning_rate": 9.597034171502256e-07,
      "loss": 0.0057,
      "reward": 1.3876937627792358,
      "reward_std": 0.16864847391843796,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3876937925815582,
      "step": 125
    },
    {
      "completion_length": 76.640625,
      "epoch": 0.08123791102514506,
      "grad_norm": 9.577360153198242,
      "kl": 0.19482421875,
      "learning_rate": 9.593810444874274e-07,
      "loss": 0.0078,
      "reward": 1.4948091506958008,
      "reward_std": 0.1558496206998825,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4948091357946396,
      "step": 126
    },
    {
      "completion_length": 78.28125,
      "epoch": 0.08188265635074146,
      "grad_norm": 136.7667236328125,
      "kl": 0.12109375,
      "learning_rate": 9.59058671824629e-07,
      "loss": 0.0048,
      "reward": 1.507809579372406,
      "reward_std": 0.21292894333600998,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.507809579372406,
      "step": 127
    },
    {
      "completion_length": 79.828125,
      "epoch": 0.08252740167633785,
      "grad_norm": 27.071666717529297,
      "kl": 0.114013671875,
      "learning_rate": 9.58736299161831e-07,
      "loss": 0.0046,
      "reward": 1.3414384722709656,
      "reward_std": 0.12231213599443436,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3414384722709656,
      "step": 128
    },
    {
      "completion_length": 75.515625,
      "epoch": 0.08317214700193423,
      "grad_norm": 18.19074058532715,
      "kl": 0.109619140625,
      "learning_rate": 9.584139264990328e-07,
      "loss": 0.0044,
      "reward": 1.5530925393104553,
      "reward_std": 0.1707591935992241,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5530924797058105,
      "step": 129
    },
    {
      "completion_length": 84.265625,
      "epoch": 0.08381689232753063,
      "grad_norm": 21.930219650268555,
      "kl": 0.104248046875,
      "learning_rate": 9.580915538362347e-07,
      "loss": 0.0042,
      "reward": 1.42109215259552,
      "reward_std": 0.2092544138431549,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.43671710789203644,
      "step": 130
    },
    {
      "completion_length": 85.125,
      "epoch": 0.08446163765312702,
      "grad_norm": 8.885394096374512,
      "kl": 0.120361328125,
      "learning_rate": 9.577691811734364e-07,
      "loss": 0.0048,
      "reward": 1.486854612827301,
      "reward_std": 0.1961679607629776,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5024795830249786,
      "step": 131
    },
    {
      "completion_length": 83.078125,
      "epoch": 0.0851063829787234,
      "grad_norm": 13.252328872680664,
      "kl": 0.200439453125,
      "learning_rate": 9.574468085106384e-07,
      "loss": 0.008,
      "reward": 1.4762264490127563,
      "reward_std": 0.16455548256635666,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4762263596057892,
      "step": 132
    },
    {
      "completion_length": 78.484375,
      "epoch": 0.08575112830431979,
      "grad_norm": 23.153955459594727,
      "kl": 0.1083984375,
      "learning_rate": 9.5712443584784e-07,
      "loss": 0.0043,
      "reward": 1.5278353691101074,
      "reward_std": 0.16748548299074173,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.527835339307785,
      "step": 133
    },
    {
      "completion_length": 86.25,
      "epoch": 0.08639587362991619,
      "grad_norm": 57.02205276489258,
      "kl": 0.11181640625,
      "learning_rate": 9.568020631850418e-07,
      "loss": 0.0045,
      "reward": 1.5299711227416992,
      "reward_std": 0.17784756422042847,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5299711525440216,
      "step": 134
    },
    {
      "completion_length": 81.5625,
      "epoch": 0.08704061895551257,
      "grad_norm": 12.712023735046387,
      "kl": 0.151611328125,
      "learning_rate": 9.564796905222435e-07,
      "loss": 0.0061,
      "reward": 1.3502278327941895,
      "reward_std": 0.13139504194259644,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3502279072999954,
      "step": 135
    },
    {
      "completion_length": 81.46875,
      "epoch": 0.08768536428110896,
      "grad_norm": 22.494131088256836,
      "kl": 0.112548828125,
      "learning_rate": 9.561573178594455e-07,
      "loss": 0.0045,
      "reward": 1.4433656334877014,
      "reward_std": 0.1883310079574585,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44336557388305664,
      "step": 136
    },
    {
      "completion_length": 82.84375,
      "epoch": 0.08833010960670536,
      "grad_norm": 16.634904861450195,
      "kl": 0.105712890625,
      "learning_rate": 9.558349451966472e-07,
      "loss": 0.0042,
      "reward": 1.470713198184967,
      "reward_std": 0.25595565140247345,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4863382577896118,
      "step": 137
    },
    {
      "completion_length": 79.125,
      "epoch": 0.08897485493230174,
      "grad_norm": 16.343263626098633,
      "kl": 0.10888671875,
      "learning_rate": 9.555125725338492e-07,
      "loss": 0.0044,
      "reward": 1.336100161075592,
      "reward_std": 0.1666717305779457,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33610019087791443,
      "step": 138
    },
    {
      "completion_length": 81.609375,
      "epoch": 0.08961960025789813,
      "grad_norm": 10.733060836791992,
      "kl": 0.10400390625,
      "learning_rate": 9.551901998710509e-07,
      "loss": 0.0042,
      "reward": 1.4843059182167053,
      "reward_std": 0.19644438475370407,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.49993085861206055,
      "step": 139
    },
    {
      "completion_length": 80.96875,
      "epoch": 0.09026434558349453,
      "grad_norm": 12.256318092346191,
      "kl": 0.114013671875,
      "learning_rate": 9.548678272082526e-07,
      "loss": 0.0046,
      "reward": 1.539031982421875,
      "reward_std": 0.21819938719272614,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5546570122241974,
      "step": 140
    },
    {
      "completion_length": 104.21875,
      "epoch": 0.09090909090909091,
      "grad_norm": 8.861552238464355,
      "kl": 0.0869140625,
      "learning_rate": 9.545454545454546e-07,
      "loss": 0.0035,
      "reward": 1.3945060968399048,
      "reward_std": 0.18161959201097488,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3945060670375824,
      "step": 141
    },
    {
      "completion_length": 90.609375,
      "epoch": 0.0915538362346873,
      "grad_norm": 16.709983825683594,
      "kl": 0.096923828125,
      "learning_rate": 9.542230818826563e-07,
      "loss": 0.0039,
      "reward": 1.5597324967384338,
      "reward_std": 0.2325654923915863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5597324520349503,
      "step": 142
    },
    {
      "completion_length": 83.5625,
      "epoch": 0.09219858156028368,
      "grad_norm": 20.069183349609375,
      "kl": 0.107666015625,
      "learning_rate": 9.53900709219858e-07,
      "loss": 0.0043,
      "reward": 1.458389699459076,
      "reward_std": 0.13381709158420563,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4583897143602371,
      "step": 143
    },
    {
      "completion_length": 83.6875,
      "epoch": 0.09284332688588008,
      "grad_norm": 82.17460632324219,
      "kl": 0.10888671875,
      "learning_rate": 9.535783365570598e-07,
      "loss": 0.0044,
      "reward": 1.3395763635635376,
      "reward_std": 0.19901933521032333,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.3552014008164406,
      "step": 144
    },
    {
      "completion_length": 88.375,
      "epoch": 0.09348807221147647,
      "grad_norm": 16.213743209838867,
      "kl": 0.09912109375,
      "learning_rate": 9.532559638942618e-07,
      "loss": 0.004,
      "reward": 1.4737032651901245,
      "reward_std": 0.20099236071109772,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4893282651901245,
      "step": 145
    },
    {
      "completion_length": 92.28125,
      "epoch": 0.09413281753707285,
      "grad_norm": 11.83709716796875,
      "kl": 0.121826171875,
      "learning_rate": 9.529335912314635e-07,
      "loss": 0.0049,
      "reward": 1.413508951663971,
      "reward_std": 0.181750126183033,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41350890696048737,
      "step": 146
    },
    {
      "completion_length": 84.640625,
      "epoch": 0.09477756286266925,
      "grad_norm": 17.66875648498535,
      "kl": 0.099609375,
      "learning_rate": 9.526112185686653e-07,
      "loss": 0.004,
      "reward": 1.4129161834716797,
      "reward_std": 0.13768525421619415,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4129162132740021,
      "step": 147
    },
    {
      "completion_length": 81.40625,
      "epoch": 0.09542230818826564,
      "grad_norm": 56.989559173583984,
      "kl": 0.112548828125,
      "learning_rate": 9.522888459058671e-07,
      "loss": 0.0045,
      "reward": 1.3882247805595398,
      "reward_std": 0.19382939487695694,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3882247507572174,
      "step": 148
    },
    {
      "completion_length": 80.625,
      "epoch": 0.09606705351386202,
      "grad_norm": 25.619226455688477,
      "kl": 0.12744140625,
      "learning_rate": 9.519664732430689e-07,
      "loss": 0.0051,
      "reward": 1.4231994152069092,
      "reward_std": 0.28173212707042694,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.45444950461387634,
      "step": 149
    },
    {
      "completion_length": 77.6875,
      "epoch": 0.09671179883945841,
      "grad_norm": 9.32150650024414,
      "kl": 0.115478515625,
      "learning_rate": 9.516441005802707e-07,
      "loss": 0.0046,
      "reward": 1.477019727230072,
      "reward_std": 0.23725295066833496,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4926447123289108,
      "step": 150
    },
    {
      "completion_length": 77.5,
      "epoch": 0.0973565441650548,
      "grad_norm": 31.08614730834961,
      "kl": 0.1240234375,
      "learning_rate": 9.513217279174726e-07,
      "loss": 0.005,
      "reward": 1.5814010500907898,
      "reward_std": 0.14891761541366577,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5814010798931122,
      "step": 151
    },
    {
      "completion_length": 79.125,
      "epoch": 0.09800128949065119,
      "grad_norm": 11.537618637084961,
      "kl": 0.119140625,
      "learning_rate": 9.509993552546743e-07,
      "loss": 0.0048,
      "reward": 1.5184857249259949,
      "reward_std": 0.1145397424697876,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5184857249259949,
      "step": 152
    },
    {
      "completion_length": 77.28125,
      "epoch": 0.09864603481624758,
      "grad_norm": 7.693099498748779,
      "kl": 0.122802734375,
      "learning_rate": 9.506769825918761e-07,
      "loss": 0.0049,
      "reward": 1.3786104917526245,
      "reward_std": 0.17887841165065765,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3786105662584305,
      "step": 153
    },
    {
      "completion_length": 75.015625,
      "epoch": 0.09929078014184398,
      "grad_norm": 16.618059158325195,
      "kl": 0.11962890625,
      "learning_rate": 9.50354609929078e-07,
      "loss": 0.0048,
      "reward": 1.597913384437561,
      "reward_std": 0.15385859459638596,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5979134142398834,
      "step": 154
    },
    {
      "completion_length": 76.71875,
      "epoch": 0.09993552546744036,
      "grad_norm": 182.9221649169922,
      "kl": 0.113037109375,
      "learning_rate": 9.500322372662798e-07,
      "loss": 0.0045,
      "reward": 1.3890660405158997,
      "reward_std": 0.13871842622756958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3890660032629967,
      "step": 155
    },
    {
      "completion_length": 75.84375,
      "epoch": 0.10058027079303675,
      "grad_norm": 16.91444969177246,
      "kl": 0.107177734375,
      "learning_rate": 9.497098646034815e-07,
      "loss": 0.0043,
      "reward": 1.5419670939445496,
      "reward_std": 0.13973253965377808,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5419670641422272,
      "step": 156
    },
    {
      "completion_length": 82.703125,
      "epoch": 0.10122501611863315,
      "grad_norm": 24.434553146362305,
      "kl": 0.1171875,
      "learning_rate": 9.493874919406834e-07,
      "loss": 0.0047,
      "reward": 1.370639443397522,
      "reward_std": 0.1973675936460495,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37063948065042496,
      "step": 157
    },
    {
      "completion_length": 82.765625,
      "epoch": 0.10186976144422953,
      "grad_norm": 25.902759552001953,
      "kl": 0.121337890625,
      "learning_rate": 9.490651192778852e-07,
      "loss": 0.0048,
      "reward": 1.4471543431282043,
      "reward_std": 0.1710544154047966,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44715429842472076,
      "step": 158
    },
    {
      "completion_length": 79.828125,
      "epoch": 0.10251450676982592,
      "grad_norm": 14.119280815124512,
      "kl": 0.10302734375,
      "learning_rate": 9.48742746615087e-07,
      "loss": 0.0041,
      "reward": 1.3451138734817505,
      "reward_std": 0.11641614139080048,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3451138883829117,
      "step": 159
    },
    {
      "completion_length": 76.1875,
      "epoch": 0.1031592520954223,
      "grad_norm": 13.816582679748535,
      "kl": 0.098876953125,
      "learning_rate": 9.484203739522888e-07,
      "loss": 0.004,
      "reward": 1.5856391787528992,
      "reward_std": 0.1286786124110222,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5856391489505768,
      "step": 160
    },
    {
      "completion_length": 77.171875,
      "epoch": 0.1038039974210187,
      "grad_norm": 8.023931503295898,
      "kl": 0.10791015625,
      "learning_rate": 9.480980012894906e-07,
      "loss": 0.0043,
      "reward": 1.607506811618805,
      "reward_std": 0.14821096509695053,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6075067818164825,
      "step": 161
    },
    {
      "completion_length": 76.65625,
      "epoch": 0.10444874274661509,
      "grad_norm": 15.326875686645508,
      "kl": 0.089111328125,
      "learning_rate": 9.477756286266923e-07,
      "loss": 0.0036,
      "reward": 1.5751556158065796,
      "reward_std": 0.16239239275455475,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5751556158065796,
      "step": 162
    },
    {
      "completion_length": 77.328125,
      "epoch": 0.10509348807221147,
      "grad_norm": 13.427696228027344,
      "kl": 0.118408203125,
      "learning_rate": 9.474532559638943e-07,
      "loss": 0.0047,
      "reward": 1.563441514968872,
      "reward_std": 0.13956208154559135,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5634415149688721,
      "step": 163
    },
    {
      "completion_length": 81.8125,
      "epoch": 0.10573823339780787,
      "grad_norm": 12.042512893676758,
      "kl": 0.109130859375,
      "learning_rate": 9.47130883301096e-07,
      "loss": 0.0044,
      "reward": 1.5153995752334595,
      "reward_std": 0.19614696502685547,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5153996497392654,
      "step": 164
    },
    {
      "completion_length": 87.84375,
      "epoch": 0.10638297872340426,
      "grad_norm": 60.0034065246582,
      "kl": 0.094970703125,
      "learning_rate": 9.468085106382978e-07,
      "loss": 0.0038,
      "reward": 1.5395575761795044,
      "reward_std": 0.1202857531607151,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.539557546377182,
      "step": 165
    },
    {
      "completion_length": 72.921875,
      "epoch": 0.10702772404900064,
      "grad_norm": 18.208284378051758,
      "kl": 0.128173828125,
      "learning_rate": 9.464861379754995e-07,
      "loss": 0.0051,
      "reward": 1.4151722192764282,
      "reward_std": 0.16611632704734802,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4307972192764282,
      "step": 166
    },
    {
      "completion_length": 74.015625,
      "epoch": 0.10767246937459704,
      "grad_norm": 19.364013671875,
      "kl": 0.100830078125,
      "learning_rate": 9.461637653127015e-07,
      "loss": 0.004,
      "reward": 1.5370397567749023,
      "reward_std": 0.1590474396944046,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5370397567749023,
      "step": 167
    },
    {
      "completion_length": 77.375,
      "epoch": 0.10831721470019343,
      "grad_norm": 10.157119750976562,
      "kl": 0.10986328125,
      "learning_rate": 9.458413926499032e-07,
      "loss": 0.0044,
      "reward": 1.3529521226882935,
      "reward_std": 0.14438863843679428,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35295212268829346,
      "step": 168
    },
    {
      "completion_length": 82.921875,
      "epoch": 0.10896196002578981,
      "grad_norm": 15.475646018981934,
      "kl": 0.126708984375,
      "learning_rate": 9.45519019987105e-07,
      "loss": 0.0051,
      "reward": 1.6346983909606934,
      "reward_std": 0.15077311173081398,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.634698361158371,
      "step": 169
    },
    {
      "completion_length": 76.34375,
      "epoch": 0.1096067053513862,
      "grad_norm": 106.46350860595703,
      "kl": 0.10498046875,
      "learning_rate": 9.451966473243068e-07,
      "loss": 0.0042,
      "reward": 1.4514729976654053,
      "reward_std": 0.12446143478155136,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45147302746772766,
      "step": 170
    },
    {
      "completion_length": 81.546875,
      "epoch": 0.1102514506769826,
      "grad_norm": 11.109757423400879,
      "kl": 0.091796875,
      "learning_rate": 9.448742746615087e-07,
      "loss": 0.0037,
      "reward": 1.4839178323745728,
      "reward_std": 0.12744403257966042,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48391780257225037,
      "step": 171
    },
    {
      "completion_length": 77.125,
      "epoch": 0.11089619600257898,
      "grad_norm": 10.060192108154297,
      "kl": 0.093994140625,
      "learning_rate": 9.445519019987105e-07,
      "loss": 0.0038,
      "reward": 1.4212124943733215,
      "reward_std": 0.17020346224308014,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4212125241756439,
      "step": 172
    },
    {
      "completion_length": 75.203125,
      "epoch": 0.11154094132817537,
      "grad_norm": 34.940547943115234,
      "kl": 0.09912109375,
      "learning_rate": 9.442295293359123e-07,
      "loss": 0.004,
      "reward": 1.5194954872131348,
      "reward_std": 0.14065240323543549,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5194954425096512,
      "step": 173
    },
    {
      "completion_length": 75.0625,
      "epoch": 0.11218568665377177,
      "grad_norm": 19.02269744873047,
      "kl": 0.099853515625,
      "learning_rate": 9.43907156673114e-07,
      "loss": 0.004,
      "reward": 1.39652419090271,
      "reward_std": 0.1206883117556572,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39652419090270996,
      "step": 174
    },
    {
      "completion_length": 74.203125,
      "epoch": 0.11283043197936815,
      "grad_norm": 10.494725227355957,
      "kl": 0.10986328125,
      "learning_rate": 9.435847840103159e-07,
      "loss": 0.0044,
      "reward": 1.5253658890724182,
      "reward_std": 0.1728542149066925,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5253658890724182,
      "step": 175
    },
    {
      "completion_length": 72.3125,
      "epoch": 0.11347517730496454,
      "grad_norm": 9.659781455993652,
      "kl": 0.10546875,
      "learning_rate": 9.432624113475178e-07,
      "loss": 0.0042,
      "reward": 1.389731228351593,
      "reward_std": 0.07970772497355938,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3897312879562378,
      "step": 176
    },
    {
      "completion_length": 70.90625,
      "epoch": 0.11411992263056092,
      "grad_norm": 14.460417747497559,
      "kl": 0.105224609375,
      "learning_rate": 9.429400386847195e-07,
      "loss": 0.0042,
      "reward": 1.516136884689331,
      "reward_std": 0.1931019425392151,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.516136884689331,
      "step": 177
    },
    {
      "completion_length": 70.484375,
      "epoch": 0.11476466795615732,
      "grad_norm": 23.858684539794922,
      "kl": 0.11328125,
      "learning_rate": 9.426176660219213e-07,
      "loss": 0.0045,
      "reward": 1.4906936287879944,
      "reward_std": 0.10323386639356613,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4906936585903168,
      "step": 178
    },
    {
      "completion_length": 75.1875,
      "epoch": 0.1154094132817537,
      "grad_norm": 11.778549194335938,
      "kl": 0.112060546875,
      "learning_rate": 9.422952933591231e-07,
      "loss": 0.0045,
      "reward": 1.5091403126716614,
      "reward_std": 0.15490031242370605,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5091403424739838,
      "step": 179
    },
    {
      "completion_length": 81.328125,
      "epoch": 0.11605415860735009,
      "grad_norm": 8.447686195373535,
      "kl": 0.1123046875,
      "learning_rate": 9.41972920696325e-07,
      "loss": 0.0045,
      "reward": 1.4372309446334839,
      "reward_std": 0.1335056535899639,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4372308552265167,
      "step": 180
    },
    {
      "completion_length": 78.96875,
      "epoch": 0.11669890393294649,
      "grad_norm": 91.26253509521484,
      "kl": 0.32568359375,
      "learning_rate": 9.416505480335267e-07,
      "loss": 0.0131,
      "reward": 1.4686675667762756,
      "reward_std": 0.16033166646957397,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46866756677627563,
      "step": 181
    },
    {
      "completion_length": 77.984375,
      "epoch": 0.11734364925854288,
      "grad_norm": 16.17279815673828,
      "kl": 0.109130859375,
      "learning_rate": 9.413281753707286e-07,
      "loss": 0.0044,
      "reward": 1.2890088558197021,
      "reward_std": 0.16506294906139374,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.28900881856679916,
      "step": 182
    },
    {
      "completion_length": 74.421875,
      "epoch": 0.11798839458413926,
      "grad_norm": 16.651771545410156,
      "kl": 0.11083984375,
      "learning_rate": 9.410058027079303e-07,
      "loss": 0.0044,
      "reward": 1.6036138534545898,
      "reward_std": 0.07109030708670616,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6036138236522675,
      "step": 183
    },
    {
      "completion_length": 82.359375,
      "epoch": 0.11863313990973566,
      "grad_norm": 14.208246231079102,
      "kl": 0.0908203125,
      "learning_rate": 9.406834300451322e-07,
      "loss": 0.0036,
      "reward": 1.5008456707000732,
      "reward_std": 0.13602961599826813,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5008457005023956,
      "step": 184
    },
    {
      "completion_length": 73.84375,
      "epoch": 0.11927788523533205,
      "grad_norm": 17.21881675720215,
      "kl": 0.098388671875,
      "learning_rate": 9.40361057382334e-07,
      "loss": 0.0039,
      "reward": 1.5844212174415588,
      "reward_std": 0.12729774415493011,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5844212472438812,
      "step": 185
    },
    {
      "completion_length": 86.96875,
      "epoch": 0.11992263056092843,
      "grad_norm": 14.471665382385254,
      "kl": 0.094970703125,
      "learning_rate": 9.400386847195358e-07,
      "loss": 0.0038,
      "reward": 1.3867006301879883,
      "reward_std": 0.1378549337387085,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3867005705833435,
      "step": 186
    },
    {
      "completion_length": 78.21875,
      "epoch": 0.12056737588652482,
      "grad_norm": 21.124332427978516,
      "kl": 0.113037109375,
      "learning_rate": 9.397163120567375e-07,
      "loss": 0.0045,
      "reward": 1.4724563956260681,
      "reward_std": 0.15684282779693604,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4724564105272293,
      "step": 187
    },
    {
      "completion_length": 74.8125,
      "epoch": 0.12121212121212122,
      "grad_norm": 16.912813186645508,
      "kl": 0.104736328125,
      "learning_rate": 9.393939393939395e-07,
      "loss": 0.0042,
      "reward": 1.4298964738845825,
      "reward_std": 0.136982262134552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4298964589834213,
      "step": 188
    },
    {
      "completion_length": 80.3125,
      "epoch": 0.1218568665377176,
      "grad_norm": 21.21564292907715,
      "kl": 0.104736328125,
      "learning_rate": 9.390715667311412e-07,
      "loss": 0.0042,
      "reward": 1.451830267906189,
      "reward_std": 0.14300556480884552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45183031260967255,
      "step": 189
    },
    {
      "completion_length": 76.140625,
      "epoch": 0.12250161186331399,
      "grad_norm": 13.693678855895996,
      "kl": 0.14208984375,
      "learning_rate": 9.38749194068343e-07,
      "loss": 0.0057,
      "reward": 1.476448655128479,
      "reward_std": 0.16905169934034348,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4764486402273178,
      "step": 190
    },
    {
      "completion_length": 98.3125,
      "epoch": 0.12314635718891039,
      "grad_norm": 12.33560562133789,
      "kl": 0.087890625,
      "learning_rate": 9.384268214055447e-07,
      "loss": 0.0035,
      "reward": 1.502742886543274,
      "reward_std": 0.17102345824241638,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.518367812037468,
      "step": 191
    },
    {
      "completion_length": 85.625,
      "epoch": 0.12379110251450677,
      "grad_norm": 17.925392150878906,
      "kl": 0.1318359375,
      "learning_rate": 9.381044487427466e-07,
      "loss": 0.0053,
      "reward": 1.5797561407089233,
      "reward_std": 0.16112536564469337,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5797560513019562,
      "step": 192
    },
    {
      "completion_length": 79.734375,
      "epoch": 0.12443584784010316,
      "grad_norm": 11.936005592346191,
      "kl": 0.109130859375,
      "learning_rate": 9.377820760799484e-07,
      "loss": 0.0044,
      "reward": 1.4955528378486633,
      "reward_std": 0.11603929847478867,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4955528676509857,
      "step": 193
    },
    {
      "completion_length": 78.375,
      "epoch": 0.12508059316569956,
      "grad_norm": 17.255386352539062,
      "kl": 0.113525390625,
      "learning_rate": 9.374597034171502e-07,
      "loss": 0.0045,
      "reward": 1.4783340692520142,
      "reward_std": 0.08274748548865318,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47833409905433655,
      "step": 194
    },
    {
      "completion_length": 84.6875,
      "epoch": 0.12572533849129594,
      "grad_norm": 25.271358489990234,
      "kl": 0.084228515625,
      "learning_rate": 9.37137330754352e-07,
      "loss": 0.0034,
      "reward": 1.3292394280433655,
      "reward_std": 0.19878728687763214,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3292394280433655,
      "step": 195
    },
    {
      "completion_length": 76.46875,
      "epoch": 0.12637008381689233,
      "grad_norm": 10.557123184204102,
      "kl": 0.10595703125,
      "learning_rate": 9.368149580915538e-07,
      "loss": 0.0042,
      "reward": 1.440163254737854,
      "reward_std": 0.17127015441656113,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4401632696390152,
      "step": 196
    },
    {
      "completion_length": 81.515625,
      "epoch": 0.1270148291424887,
      "grad_norm": 27.85223960876465,
      "kl": 0.11083984375,
      "learning_rate": 9.364925854287556e-07,
      "loss": 0.0044,
      "reward": 1.3243326544761658,
      "reward_std": 0.10718319937586784,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3243326246738434,
      "step": 197
    },
    {
      "completion_length": 86.25,
      "epoch": 0.1276595744680851,
      "grad_norm": 30.643444061279297,
      "kl": 0.090576171875,
      "learning_rate": 9.361702127659575e-07,
      "loss": 0.0036,
      "reward": 1.4653456211090088,
      "reward_std": 0.17089767009019852,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4653455317020416,
      "step": 198
    },
    {
      "completion_length": 89.03125,
      "epoch": 0.1283043197936815,
      "grad_norm": 10.595077514648438,
      "kl": 0.10888671875,
      "learning_rate": 9.358478401031592e-07,
      "loss": 0.0044,
      "reward": 1.3735358119010925,
      "reward_std": 0.17685279250144958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37353579699993134,
      "step": 199
    },
    {
      "completion_length": 87.375,
      "epoch": 0.1289490651192779,
      "grad_norm": 17.84724235534668,
      "kl": 0.1533203125,
      "learning_rate": 9.35525467440361e-07,
      "loss": 0.0061,
      "reward": 1.4756492972373962,
      "reward_std": 0.09536949545145035,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47564925253391266,
      "step": 200
    },
    {
      "completion_length": 79.09375,
      "epoch": 0.12959381044487428,
      "grad_norm": 13.18930435180664,
      "kl": 0.0927734375,
      "learning_rate": 9.352030947775629e-07,
      "loss": 0.0037,
      "reward": 1.512624442577362,
      "reward_std": 0.10898591205477715,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5126243680715561,
      "step": 201
    },
    {
      "completion_length": 82.21875,
      "epoch": 0.13023855577047067,
      "grad_norm": 20.932710647583008,
      "kl": 0.123046875,
      "learning_rate": 9.348807221147647e-07,
      "loss": 0.0049,
      "reward": 1.4443880915641785,
      "reward_std": 0.08235402405261993,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44438812136650085,
      "step": 202
    },
    {
      "completion_length": 82.234375,
      "epoch": 0.13088330109606705,
      "grad_norm": 10.726685523986816,
      "kl": 0.108642578125,
      "learning_rate": 9.345583494519664e-07,
      "loss": 0.0043,
      "reward": 1.6250720024108887,
      "reward_std": 0.12261708825826645,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6250719428062439,
      "step": 203
    },
    {
      "completion_length": 82.6875,
      "epoch": 0.13152804642166344,
      "grad_norm": 10.96521282196045,
      "kl": 0.1015625,
      "learning_rate": 9.342359767891683e-07,
      "loss": 0.0041,
      "reward": 1.6589622497558594,
      "reward_std": 0.13084128499031067,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6589623093605042,
      "step": 204
    },
    {
      "completion_length": 80.921875,
      "epoch": 0.13217279174725982,
      "grad_norm": 18.59563446044922,
      "kl": 0.1103515625,
      "learning_rate": 9.3391360412637e-07,
      "loss": 0.0044,
      "reward": 1.4989012479782104,
      "reward_std": 0.12866604328155518,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49890124797821045,
      "step": 205
    },
    {
      "completion_length": 79.140625,
      "epoch": 0.13281753707285623,
      "grad_norm": 26.62146759033203,
      "kl": 0.11962890625,
      "learning_rate": 9.335912314635719e-07,
      "loss": 0.0048,
      "reward": 1.557816207408905,
      "reward_std": 0.19459083676338196,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5578161776065826,
      "step": 206
    },
    {
      "completion_length": 87.34375,
      "epoch": 0.13346228239845262,
      "grad_norm": 15.918787956237793,
      "kl": 0.1044921875,
      "learning_rate": 9.332688588007737e-07,
      "loss": 0.0042,
      "reward": 1.459834098815918,
      "reward_std": 0.1547633334994316,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4598340541124344,
      "step": 207
    },
    {
      "completion_length": 92.734375,
      "epoch": 0.134107027724049,
      "grad_norm": 9.844905853271484,
      "kl": 0.110107421875,
      "learning_rate": 9.329464861379755e-07,
      "loss": 0.0044,
      "reward": 1.5493760108947754,
      "reward_std": 0.13573891669511795,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5493760406970978,
      "step": 208
    },
    {
      "completion_length": 82.203125,
      "epoch": 0.1347517730496454,
      "grad_norm": 9.640893936157227,
      "kl": 0.1474609375,
      "learning_rate": 9.326241134751772e-07,
      "loss": 0.0059,
      "reward": 1.4705365896224976,
      "reward_std": 0.117528036236763,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47053661942481995,
      "step": 209
    },
    {
      "completion_length": 84.71875,
      "epoch": 0.13539651837524178,
      "grad_norm": 20.125877380371094,
      "kl": 0.11279296875,
      "learning_rate": 9.323017408123792e-07,
      "loss": 0.0045,
      "reward": 1.5758726596832275,
      "reward_std": 0.12206336110830307,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5758726894855499,
      "step": 210
    },
    {
      "completion_length": 82.65625,
      "epoch": 0.13604126370083816,
      "grad_norm": 12.198143005371094,
      "kl": 0.13134765625,
      "learning_rate": 9.319793681495809e-07,
      "loss": 0.0052,
      "reward": 1.5487463474273682,
      "reward_std": 0.16695217788219452,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5487463772296906,
      "step": 211
    },
    {
      "completion_length": 86.25,
      "epoch": 0.13668600902643455,
      "grad_norm": 12.829062461853027,
      "kl": 0.111083984375,
      "learning_rate": 9.316569954867827e-07,
      "loss": 0.0044,
      "reward": 1.6886752843856812,
      "reward_std": 0.10757141932845116,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6886752545833588,
      "step": 212
    },
    {
      "completion_length": 79.265625,
      "epoch": 0.13733075435203096,
      "grad_norm": 25.454059600830078,
      "kl": 0.12109375,
      "learning_rate": 9.313346228239844e-07,
      "loss": 0.0048,
      "reward": 1.6044875383377075,
      "reward_std": 0.11038105189800262,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6044874787330627,
      "step": 213
    },
    {
      "completion_length": 72.921875,
      "epoch": 0.13797549967762734,
      "grad_norm": 19.579898834228516,
      "kl": 0.150390625,
      "learning_rate": 9.310122501611864e-07,
      "loss": 0.006,
      "reward": 1.3529685735702515,
      "reward_std": 0.06810446828603745,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35296855866909027,
      "step": 214
    },
    {
      "completion_length": 75.328125,
      "epoch": 0.13862024500322373,
      "grad_norm": 20.17354393005371,
      "kl": 0.110107421875,
      "learning_rate": 9.306898774983881e-07,
      "loss": 0.0044,
      "reward": 1.5281617641448975,
      "reward_std": 0.10443935543298721,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5281617343425751,
      "step": 215
    },
    {
      "completion_length": 79.3125,
      "epoch": 0.13926499032882012,
      "grad_norm": 18.318899154663086,
      "kl": 0.125,
      "learning_rate": 9.303675048355899e-07,
      "loss": 0.005,
      "reward": 1.4097517132759094,
      "reward_std": 0.12294080853462219,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.409751757979393,
      "step": 216
    },
    {
      "completion_length": 75.828125,
      "epoch": 0.1399097356544165,
      "grad_norm": 16.73103141784668,
      "kl": 0.14013671875,
      "learning_rate": 9.300451321727917e-07,
      "loss": 0.0056,
      "reward": 1.5454647541046143,
      "reward_std": 0.15932783484458923,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.545464813709259,
      "step": 217
    },
    {
      "completion_length": 77.5625,
      "epoch": 0.14055448098001289,
      "grad_norm": 16.411338806152344,
      "kl": 0.1240234375,
      "learning_rate": 9.297227595099935e-07,
      "loss": 0.005,
      "reward": 1.4458375573158264,
      "reward_std": 0.10837087035179138,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4458375424146652,
      "step": 218
    },
    {
      "completion_length": 80.828125,
      "epoch": 0.14119922630560927,
      "grad_norm": 18.790870666503906,
      "kl": 0.12255859375,
      "learning_rate": 9.294003868471953e-07,
      "loss": 0.0049,
      "reward": 1.6049268245697021,
      "reward_std": 0.15314558148384094,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6049268841743469,
      "step": 219
    },
    {
      "completion_length": 83.234375,
      "epoch": 0.14184397163120568,
      "grad_norm": 24.741748809814453,
      "kl": 0.114501953125,
      "learning_rate": 9.290780141843972e-07,
      "loss": 0.0046,
      "reward": 1.4110137820243835,
      "reward_std": 0.11946087330579758,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4110138416290283,
      "step": 220
    },
    {
      "completion_length": 77.90625,
      "epoch": 0.14248871695680207,
      "grad_norm": 17.31515884399414,
      "kl": 0.13330078125,
      "learning_rate": 9.287556415215989e-07,
      "loss": 0.0053,
      "reward": 1.4192990064620972,
      "reward_std": 0.13341504707932472,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4192989766597748,
      "step": 221
    },
    {
      "completion_length": 75.484375,
      "epoch": 0.14313346228239845,
      "grad_norm": 17.7659969329834,
      "kl": 0.118896484375,
      "learning_rate": 9.284332688588007e-07,
      "loss": 0.0048,
      "reward": 1.6054986715316772,
      "reward_std": 0.11470058560371399,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6054986715316772,
      "step": 222
    },
    {
      "completion_length": 77.8125,
      "epoch": 0.14377820760799484,
      "grad_norm": 17.46185302734375,
      "kl": 0.13720703125,
      "learning_rate": 9.281108961960026e-07,
      "loss": 0.0055,
      "reward": 1.4767987728118896,
      "reward_std": 0.20763688534498215,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47679875791072845,
      "step": 223
    },
    {
      "completion_length": 73.109375,
      "epoch": 0.14442295293359123,
      "grad_norm": 11.976152420043945,
      "kl": 0.137939453125,
      "learning_rate": 9.277885235332044e-07,
      "loss": 0.0055,
      "reward": 1.5722563862800598,
      "reward_std": 0.13132376223802567,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.572256326675415,
      "step": 224
    },
    {
      "completion_length": 78.484375,
      "epoch": 0.1450676982591876,
      "grad_norm": 9.927757263183594,
      "kl": 0.130859375,
      "learning_rate": 9.274661508704061e-07,
      "loss": 0.0052,
      "reward": 1.5556007027626038,
      "reward_std": 0.154083713889122,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.555600643157959,
      "step": 225
    },
    {
      "completion_length": 79.609375,
      "epoch": 0.14571244358478402,
      "grad_norm": 13.89825439453125,
      "kl": 0.146484375,
      "learning_rate": 9.27143778207608e-07,
      "loss": 0.0059,
      "reward": 1.6433384418487549,
      "reward_std": 0.14611933380365372,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6433383822441101,
      "step": 226
    },
    {
      "completion_length": 89.265625,
      "epoch": 0.1463571889103804,
      "grad_norm": 11.447097778320312,
      "kl": 0.1240234375,
      "learning_rate": 9.268214055448098e-07,
      "loss": 0.005,
      "reward": 1.4545153975486755,
      "reward_std": 0.14470210671424866,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4545154720544815,
      "step": 227
    },
    {
      "completion_length": 97.6875,
      "epoch": 0.1470019342359768,
      "grad_norm": 17.05489730834961,
      "kl": 0.122314453125,
      "learning_rate": 9.264990328820116e-07,
      "loss": 0.0049,
      "reward": 1.570904791355133,
      "reward_std": 0.11560512334108353,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5709047913551331,
      "step": 228
    },
    {
      "completion_length": 86.703125,
      "epoch": 0.14764667956157318,
      "grad_norm": 13.748475074768066,
      "kl": 0.103271484375,
      "learning_rate": 9.261766602192134e-07,
      "loss": 0.0041,
      "reward": 1.6213688850402832,
      "reward_std": 0.09734684601426125,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6213689148426056,
      "step": 229
    },
    {
      "completion_length": 81.921875,
      "epoch": 0.14829142488716957,
      "grad_norm": 16.89826774597168,
      "kl": 0.13720703125,
      "learning_rate": 9.258542875564152e-07,
      "loss": 0.0055,
      "reward": 1.439545214176178,
      "reward_std": 0.1173011064529419,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4395451992750168,
      "step": 230
    },
    {
      "completion_length": 90.84375,
      "epoch": 0.14893617021276595,
      "grad_norm": 13.908186912536621,
      "kl": 0.09033203125,
      "learning_rate": 9.255319148936169e-07,
      "loss": 0.0036,
      "reward": 1.4615440964698792,
      "reward_std": 0.17866834998130798,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46154411137104034,
      "step": 231
    },
    {
      "completion_length": 93.265625,
      "epoch": 0.14958091553836234,
      "grad_norm": 12.506692886352539,
      "kl": 0.09814453125,
      "learning_rate": 9.252095422308189e-07,
      "loss": 0.0039,
      "reward": 1.508992850780487,
      "reward_std": 0.10917293652892113,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5089928805828094,
      "step": 232
    },
    {
      "completion_length": 99.0625,
      "epoch": 0.15022566086395875,
      "grad_norm": 28.037565231323242,
      "kl": 0.12109375,
      "learning_rate": 9.248871695680206e-07,
      "loss": 0.0048,
      "reward": 1.5570703744888306,
      "reward_std": 0.13100450485944748,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5570703744888306,
      "step": 233
    },
    {
      "completion_length": 98.75,
      "epoch": 0.15087040618955513,
      "grad_norm": 18.369422912597656,
      "kl": 0.13671875,
      "learning_rate": 9.245647969052224e-07,
      "loss": 0.0055,
      "reward": 1.5361040830612183,
      "reward_std": 0.14577998965978622,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5361041128635406,
      "step": 234
    },
    {
      "completion_length": 102.828125,
      "epoch": 0.15151515151515152,
      "grad_norm": 22.57931137084961,
      "kl": 0.110595703125,
      "learning_rate": 9.242424242424241e-07,
      "loss": 0.0044,
      "reward": 1.6102051734924316,
      "reward_std": 0.13294027745723724,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6102051734924316,
      "step": 235
    },
    {
      "completion_length": 99.640625,
      "epoch": 0.1521598968407479,
      "grad_norm": 6.480432510375977,
      "kl": 0.106689453125,
      "learning_rate": 9.239200515796261e-07,
      "loss": 0.0043,
      "reward": 1.5354961156845093,
      "reward_std": 0.12877590954303741,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5354961007833481,
      "step": 236
    },
    {
      "completion_length": 104.828125,
      "epoch": 0.1528046421663443,
      "grad_norm": 9.36783504486084,
      "kl": 0.10791015625,
      "learning_rate": 9.235976789168278e-07,
      "loss": 0.0043,
      "reward": 1.506102979183197,
      "reward_std": 0.14209586381912231,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5061030089855194,
      "step": 237
    },
    {
      "completion_length": 101.828125,
      "epoch": 0.15344938749194068,
      "grad_norm": 12.477219581604004,
      "kl": 0.099853515625,
      "learning_rate": 9.232753062540296e-07,
      "loss": 0.004,
      "reward": 1.4833937287330627,
      "reward_std": 0.1376461237668991,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4833938032388687,
      "step": 238
    },
    {
      "completion_length": 98.625,
      "epoch": 0.15409413281753706,
      "grad_norm": 24.562990188598633,
      "kl": 0.1083984375,
      "learning_rate": 9.229529335912314e-07,
      "loss": 0.0043,
      "reward": 1.4200094938278198,
      "reward_std": 0.1750444434583187,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42000944912433624,
      "step": 239
    },
    {
      "completion_length": 106.125,
      "epoch": 0.15473887814313347,
      "grad_norm": 10.694780349731445,
      "kl": 0.1337890625,
      "learning_rate": 9.226305609284333e-07,
      "loss": 0.0054,
      "reward": 1.5045281648635864,
      "reward_std": 0.12367697432637215,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5045281946659088,
      "step": 240
    },
    {
      "completion_length": 100.09375,
      "epoch": 0.15538362346872986,
      "grad_norm": 14.116943359375,
      "kl": 0.108642578125,
      "learning_rate": 9.22308188265635e-07,
      "loss": 0.0043,
      "reward": 1.593424916267395,
      "reward_std": 0.1556602418422699,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5934248268604279,
      "step": 241
    },
    {
      "completion_length": 105.921875,
      "epoch": 0.15602836879432624,
      "grad_norm": 17.899860382080078,
      "kl": 0.12109375,
      "learning_rate": 9.219858156028369e-07,
      "loss": 0.0048,
      "reward": 1.5431109070777893,
      "reward_std": 0.1369476467370987,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5431108772754669,
      "step": 242
    },
    {
      "completion_length": 113.890625,
      "epoch": 0.15667311411992263,
      "grad_norm": 13.18708610534668,
      "kl": 0.09130859375,
      "learning_rate": 9.216634429400386e-07,
      "loss": 0.0037,
      "reward": 1.470144808292389,
      "reward_std": 0.13966727256774902,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47014474868774414,
      "step": 243
    },
    {
      "completion_length": 99.796875,
      "epoch": 0.15731785944551901,
      "grad_norm": 7.642358303070068,
      "kl": 0.158203125,
      "learning_rate": 9.213410702772404e-07,
      "loss": 0.0063,
      "reward": 1.3841314911842346,
      "reward_std": 0.11086063086986542,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.38413146138191223,
      "step": 244
    },
    {
      "completion_length": 108.578125,
      "epoch": 0.1579626047711154,
      "grad_norm": 16.818422317504883,
      "kl": 0.09423828125,
      "learning_rate": 9.210186976144423e-07,
      "loss": 0.0038,
      "reward": 1.4686157703399658,
      "reward_std": 0.09336278215050697,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4686158001422882,
      "step": 245
    },
    {
      "completion_length": 103.609375,
      "epoch": 0.15860735009671179,
      "grad_norm": 21.795143127441406,
      "kl": 0.1044921875,
      "learning_rate": 9.206963249516441e-07,
      "loss": 0.0042,
      "reward": 1.3981711864471436,
      "reward_std": 0.20313138514757156,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39817118644714355,
      "step": 246
    },
    {
      "completion_length": 109.046875,
      "epoch": 0.1592520954223082,
      "grad_norm": 17.059049606323242,
      "kl": 0.09765625,
      "learning_rate": 9.203739522888458e-07,
      "loss": 0.0039,
      "reward": 1.5794776678085327,
      "reward_std": 0.16649839282035828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5794776380062103,
      "step": 247
    },
    {
      "completion_length": 99.296875,
      "epoch": 0.15989684074790458,
      "grad_norm": 12.510836601257324,
      "kl": 0.1064453125,
      "learning_rate": 9.200515796260477e-07,
      "loss": 0.0043,
      "reward": 1.6266114115715027,
      "reward_std": 0.16728466749191284,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6266114115715027,
      "step": 248
    },
    {
      "completion_length": 95.390625,
      "epoch": 0.16054158607350097,
      "grad_norm": 21.999820709228516,
      "kl": 0.10498046875,
      "learning_rate": 9.197292069632495e-07,
      "loss": 0.0042,
      "reward": 1.7057658433914185,
      "reward_std": 0.15626183152198792,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7057657837867737,
      "step": 249
    },
    {
      "completion_length": 89.6875,
      "epoch": 0.16118633139909735,
      "grad_norm": 12.472075462341309,
      "kl": 0.094482421875,
      "learning_rate": 9.194068343004513e-07,
      "loss": 0.0038,
      "reward": 1.5567827224731445,
      "reward_std": 0.1044660434126854,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5567827373743057,
      "step": 250
    },
    {
      "completion_length": 99.546875,
      "epoch": 0.16183107672469374,
      "grad_norm": 10.50654125213623,
      "kl": 0.09619140625,
      "learning_rate": 9.19084461637653e-07,
      "loss": 0.0038,
      "reward": 1.5309272408485413,
      "reward_std": 0.12022281438112259,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5309272557497025,
      "step": 251
    },
    {
      "completion_length": 113.09375,
      "epoch": 0.16247582205029013,
      "grad_norm": 9.103067398071289,
      "kl": 0.0927734375,
      "learning_rate": 9.187620889748549e-07,
      "loss": 0.0037,
      "reward": 1.5212771892547607,
      "reward_std": 0.13080750405788422,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5212771892547607,
      "step": 252
    },
    {
      "completion_length": 101.203125,
      "epoch": 0.16312056737588654,
      "grad_norm": 11.426836013793945,
      "kl": 0.105224609375,
      "learning_rate": 9.184397163120567e-07,
      "loss": 0.0042,
      "reward": 1.569593608379364,
      "reward_std": 0.14871494472026825,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5695934891700745,
      "step": 253
    },
    {
      "completion_length": 98.203125,
      "epoch": 0.16376531270148292,
      "grad_norm": 9.3218355178833,
      "kl": 0.090576171875,
      "learning_rate": 9.181173436492586e-07,
      "loss": 0.0036,
      "reward": 1.530907690525055,
      "reward_std": 0.0898456946015358,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5309076905250549,
      "step": 254
    },
    {
      "completion_length": 100.234375,
      "epoch": 0.1644100580270793,
      "grad_norm": 12.731607437133789,
      "kl": 0.076171875,
      "learning_rate": 9.177949709864603e-07,
      "loss": 0.0031,
      "reward": 1.4146405458450317,
      "reward_std": 0.11831414513289928,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41464051604270935,
      "step": 255
    },
    {
      "completion_length": 93.109375,
      "epoch": 0.1650548033526757,
      "grad_norm": 17.456554412841797,
      "kl": 0.098876953125,
      "learning_rate": 9.174725983236621e-07,
      "loss": 0.004,
      "reward": 1.5557602643966675,
      "reward_std": 0.11968041211366653,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5557602941989899,
      "step": 256
    },
    {
      "completion_length": 100.140625,
      "epoch": 0.16569954867827208,
      "grad_norm": 12.351179122924805,
      "kl": 0.109375,
      "learning_rate": 9.171502256608638e-07,
      "loss": 0.0044,
      "reward": 1.5976706147193909,
      "reward_std": 0.1560915857553482,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5976706445217133,
      "step": 257
    },
    {
      "completion_length": 101.890625,
      "epoch": 0.16634429400386846,
      "grad_norm": 12.943246841430664,
      "kl": 0.089599609375,
      "learning_rate": 9.168278529980658e-07,
      "loss": 0.0036,
      "reward": 1.3898320198059082,
      "reward_std": 0.12752876058220863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.389832004904747,
      "step": 258
    },
    {
      "completion_length": 92.40625,
      "epoch": 0.16698903932946485,
      "grad_norm": 18.802291870117188,
      "kl": 0.094970703125,
      "learning_rate": 9.165054803352675e-07,
      "loss": 0.0038,
      "reward": 1.5527189373970032,
      "reward_std": 0.1755797415971756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.552718997001648,
      "step": 259
    },
    {
      "completion_length": 92.9375,
      "epoch": 0.16763378465506126,
      "grad_norm": 22.909547805786133,
      "kl": 0.1005859375,
      "learning_rate": 9.161831076724693e-07,
      "loss": 0.004,
      "reward": 1.3774397373199463,
      "reward_std": 0.17273323237895966,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3774396479129791,
      "step": 260
    },
    {
      "completion_length": 94.609375,
      "epoch": 0.16827852998065765,
      "grad_norm": 9.854948997497559,
      "kl": 0.088623046875,
      "learning_rate": 9.158607350096711e-07,
      "loss": 0.0035,
      "reward": 1.5603925585746765,
      "reward_std": 0.14747831225395203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5603925287723541,
      "step": 261
    },
    {
      "completion_length": 104.125,
      "epoch": 0.16892327530625403,
      "grad_norm": 12.83651351928711,
      "kl": 0.088134765625,
      "learning_rate": 9.15538362346873e-07,
      "loss": 0.0035,
      "reward": 1.5412468910217285,
      "reward_std": 0.14012236148118973,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5412469208240509,
      "step": 262
    },
    {
      "completion_length": 97.625,
      "epoch": 0.16956802063185042,
      "grad_norm": 11.614677429199219,
      "kl": 0.095703125,
      "learning_rate": 9.152159896840747e-07,
      "loss": 0.0038,
      "reward": 1.3001367449760437,
      "reward_std": 0.1370644010603428,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3001367151737213,
      "step": 263
    },
    {
      "completion_length": 97.625,
      "epoch": 0.1702127659574468,
      "grad_norm": 10.85827350616455,
      "kl": 0.107666015625,
      "learning_rate": 9.148936170212766e-07,
      "loss": 0.0043,
      "reward": 1.5808395147323608,
      "reward_std": 0.1318642795085907,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5808394998311996,
      "step": 264
    },
    {
      "completion_length": 94.0,
      "epoch": 0.1708575112830432,
      "grad_norm": 19.177623748779297,
      "kl": 0.0869140625,
      "learning_rate": 9.145712443584783e-07,
      "loss": 0.0035,
      "reward": 1.428877592086792,
      "reward_std": 0.1186145581305027,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.428877592086792,
      "step": 265
    },
    {
      "completion_length": 101.171875,
      "epoch": 0.17150225660863957,
      "grad_norm": 14.279739379882812,
      "kl": 0.1953125,
      "learning_rate": 9.142488716956802e-07,
      "loss": 0.0078,
      "reward": 1.5931092500686646,
      "reward_std": 0.1707104742527008,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5931092202663422,
      "step": 266
    },
    {
      "completion_length": 101.0,
      "epoch": 0.172147001934236,
      "grad_norm": 11.718372344970703,
      "kl": 0.085693359375,
      "learning_rate": 9.13926499032882e-07,
      "loss": 0.0034,
      "reward": 1.5582287907600403,
      "reward_std": 0.10395102947950363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5582287758588791,
      "step": 267
    },
    {
      "completion_length": 99.078125,
      "epoch": 0.17279174725983237,
      "grad_norm": 11.058600425720215,
      "kl": 0.092529296875,
      "learning_rate": 9.136041263700838e-07,
      "loss": 0.0037,
      "reward": 1.4419950246810913,
      "reward_std": 0.1245020180940628,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4419950246810913,
      "step": 268
    },
    {
      "completion_length": 98.3125,
      "epoch": 0.17343649258542876,
      "grad_norm": 18.12723731994629,
      "kl": 0.15478515625,
      "learning_rate": 9.132817537072855e-07,
      "loss": 0.0062,
      "reward": 1.470383882522583,
      "reward_std": 0.21054161339998245,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4703838974237442,
      "step": 269
    },
    {
      "completion_length": 104.53125,
      "epoch": 0.17408123791102514,
      "grad_norm": 10.150888442993164,
      "kl": 0.101806640625,
      "learning_rate": 9.129593810444874e-07,
      "loss": 0.0041,
      "reward": 1.4431976079940796,
      "reward_std": 0.20065002888441086,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4431975781917572,
      "step": 270
    },
    {
      "completion_length": 109.90625,
      "epoch": 0.17472598323662153,
      "grad_norm": 7.890026569366455,
      "kl": 0.082275390625,
      "learning_rate": 9.126370083816892e-07,
      "loss": 0.0033,
      "reward": 1.4148862957954407,
      "reward_std": 0.08163053542375565,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4148862734436989,
      "step": 271
    },
    {
      "completion_length": 95.640625,
      "epoch": 0.17537072856221791,
      "grad_norm": 10.980069160461426,
      "kl": 0.099853515625,
      "learning_rate": 9.12314635718891e-07,
      "loss": 0.004,
      "reward": 1.4476662278175354,
      "reward_std": 0.17421916499733925,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.447666272521019,
      "step": 272
    },
    {
      "completion_length": 107.625,
      "epoch": 0.1760154738878143,
      "grad_norm": 20.894821166992188,
      "kl": 0.0830078125,
      "learning_rate": 9.119922630560928e-07,
      "loss": 0.0033,
      "reward": 1.5147703289985657,
      "reward_std": 0.11893308907747269,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5147703588008881,
      "step": 273
    },
    {
      "completion_length": 97.5625,
      "epoch": 0.1766602192134107,
      "grad_norm": 17.869699478149414,
      "kl": 0.1171875,
      "learning_rate": 9.116698903932946e-07,
      "loss": 0.0047,
      "reward": 1.5422923564910889,
      "reward_std": 0.15741287916898727,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5422923564910889,
      "step": 274
    },
    {
      "completion_length": 95.734375,
      "epoch": 0.1773049645390071,
      "grad_norm": 15.127033233642578,
      "kl": 0.11328125,
      "learning_rate": 9.113475177304964e-07,
      "loss": 0.0045,
      "reward": 1.4402891397476196,
      "reward_std": 0.12576143443584442,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44028913974761963,
      "step": 275
    },
    {
      "completion_length": 116.8125,
      "epoch": 0.17794970986460348,
      "grad_norm": 31.529743194580078,
      "kl": 0.087890625,
      "learning_rate": 9.110251450676983e-07,
      "loss": 0.0035,
      "reward": 1.5826689004898071,
      "reward_std": 0.17237188667058945,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5826689004898071,
      "step": 276
    },
    {
      "completion_length": 99.359375,
      "epoch": 0.17859445519019987,
      "grad_norm": 24.593116760253906,
      "kl": 0.09521484375,
      "learning_rate": 9.107027724049e-07,
      "loss": 0.0038,
      "reward": 1.4288339614868164,
      "reward_std": 0.13205820694565773,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42883390188217163,
      "step": 277
    },
    {
      "completion_length": 108.5,
      "epoch": 0.17923920051579625,
      "grad_norm": 16.80621337890625,
      "kl": 0.090087890625,
      "learning_rate": 9.103803997421018e-07,
      "loss": 0.0036,
      "reward": 1.4728108644485474,
      "reward_std": 0.14833205938339233,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4728108048439026,
      "step": 278
    },
    {
      "completion_length": 99.421875,
      "epoch": 0.17988394584139264,
      "grad_norm": 11.30988597869873,
      "kl": 0.095947265625,
      "learning_rate": 9.100580270793036e-07,
      "loss": 0.0038,
      "reward": 1.5478020310401917,
      "reward_std": 0.10750558972358704,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5478019714355469,
      "step": 279
    },
    {
      "completion_length": 111.9375,
      "epoch": 0.18052869116698905,
      "grad_norm": 18.844890594482422,
      "kl": 0.091552734375,
      "learning_rate": 9.097356544165055e-07,
      "loss": 0.0037,
      "reward": 1.5706157684326172,
      "reward_std": 0.13911627978086472,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5706157982349396,
      "step": 280
    },
    {
      "completion_length": 115.703125,
      "epoch": 0.18117343649258544,
      "grad_norm": 58.691158294677734,
      "kl": 0.105712890625,
      "learning_rate": 9.094132817537072e-07,
      "loss": 0.0042,
      "reward": 1.4865358471870422,
      "reward_std": 0.11797124892473221,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48653586208820343,
      "step": 281
    },
    {
      "completion_length": 105.0625,
      "epoch": 0.18181818181818182,
      "grad_norm": 21.599842071533203,
      "kl": 0.096435546875,
      "learning_rate": 9.09090909090909e-07,
      "loss": 0.0039,
      "reward": 1.439337968826294,
      "reward_std": 0.13826286792755127,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43933798372745514,
      "step": 282
    },
    {
      "completion_length": 111.734375,
      "epoch": 0.1824629271437782,
      "grad_norm": 9.935829162597656,
      "kl": 0.08740234375,
      "learning_rate": 9.087685364281109e-07,
      "loss": 0.0035,
      "reward": 1.5908172130584717,
      "reward_std": 0.11852175369858742,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5908172726631165,
      "step": 283
    },
    {
      "completion_length": 110.484375,
      "epoch": 0.1831076724693746,
      "grad_norm": 10.622421264648438,
      "kl": 0.10888671875,
      "learning_rate": 9.084461637653127e-07,
      "loss": 0.0044,
      "reward": 1.4755669236183167,
      "reward_std": 0.12218554317951202,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47556689381599426,
      "step": 284
    },
    {
      "completion_length": 115.5625,
      "epoch": 0.18375241779497098,
      "grad_norm": 6.261775493621826,
      "kl": 0.090576171875,
      "learning_rate": 9.081237911025144e-07,
      "loss": 0.0036,
      "reward": 1.5844565033912659,
      "reward_std": 0.09432502463459969,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5844565629959106,
      "step": 285
    },
    {
      "completion_length": 109.703125,
      "epoch": 0.18439716312056736,
      "grad_norm": 15.250070571899414,
      "kl": 0.109375,
      "learning_rate": 9.078014184397163e-07,
      "loss": 0.0044,
      "reward": 1.5829523801803589,
      "reward_std": 0.15015285462141037,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5829523801803589,
      "step": 286
    },
    {
      "completion_length": 108.5,
      "epoch": 0.18504190844616378,
      "grad_norm": 8.247808456420898,
      "kl": 0.087890625,
      "learning_rate": 9.07479045776918e-07,
      "loss": 0.0035,
      "reward": 1.5334210991859436,
      "reward_std": 0.15494172275066376,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5334210991859436,
      "step": 287
    },
    {
      "completion_length": 111.765625,
      "epoch": 0.18568665377176016,
      "grad_norm": 45.70930099487305,
      "kl": 0.227783203125,
      "learning_rate": 9.071566731141199e-07,
      "loss": 0.0091,
      "reward": 1.478054404258728,
      "reward_std": 0.1481890268623829,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4780544489622116,
      "step": 288
    },
    {
      "completion_length": 105.546875,
      "epoch": 0.18633139909735655,
      "grad_norm": 11.30000114440918,
      "kl": 0.1005859375,
      "learning_rate": 9.068343004513217e-07,
      "loss": 0.004,
      "reward": 1.5877296924591064,
      "reward_std": 0.1937682181596756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5877296328544617,
      "step": 289
    },
    {
      "completion_length": 103.25,
      "epoch": 0.18697614442295293,
      "grad_norm": 10.429908752441406,
      "kl": 0.166259765625,
      "learning_rate": 9.065119277885235e-07,
      "loss": 0.0067,
      "reward": 1.6546611785888672,
      "reward_std": 0.15019819140434265,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6546612083911896,
      "step": 290
    },
    {
      "completion_length": 101.984375,
      "epoch": 0.18762088974854932,
      "grad_norm": 21.050172805786133,
      "kl": 0.1064453125,
      "learning_rate": 9.061895551257252e-07,
      "loss": 0.0043,
      "reward": 1.5703359842300415,
      "reward_std": 0.1420222818851471,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5703359246253967,
      "step": 291
    },
    {
      "completion_length": 95.890625,
      "epoch": 0.1882656350741457,
      "grad_norm": 9.279730796813965,
      "kl": 0.11181640625,
      "learning_rate": 9.058671824629272e-07,
      "loss": 0.0045,
      "reward": 1.480491280555725,
      "reward_std": 0.11640884727239609,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4804912358522415,
      "step": 292
    },
    {
      "completion_length": 104.796875,
      "epoch": 0.1889103803997421,
      "grad_norm": 13.045880317687988,
      "kl": 0.099365234375,
      "learning_rate": 9.055448098001289e-07,
      "loss": 0.004,
      "reward": 1.54798024892807,
      "reward_std": 0.15294693410396576,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5479801893234253,
      "step": 293
    },
    {
      "completion_length": 93.859375,
      "epoch": 0.1895551257253385,
      "grad_norm": 9.018101692199707,
      "kl": 0.104736328125,
      "learning_rate": 9.052224371373307e-07,
      "loss": 0.0042,
      "reward": 1.5038177371025085,
      "reward_std": 0.10080670937895775,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5038177520036697,
      "step": 294
    },
    {
      "completion_length": 107.171875,
      "epoch": 0.1901998710509349,
      "grad_norm": 8.838314056396484,
      "kl": 0.0869140625,
      "learning_rate": 9.049000644745325e-07,
      "loss": 0.0035,
      "reward": 1.5588614344596863,
      "reward_std": 0.11014912649989128,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5588614046573639,
      "step": 295
    },
    {
      "completion_length": 99.875,
      "epoch": 0.19084461637653127,
      "grad_norm": 12.840435981750488,
      "kl": 0.097900390625,
      "learning_rate": 9.045776918117344e-07,
      "loss": 0.0039,
      "reward": 1.5545161366462708,
      "reward_std": 0.12117831408977509,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5545161068439484,
      "step": 296
    },
    {
      "completion_length": 97.921875,
      "epoch": 0.19148936170212766,
      "grad_norm": 14.086722373962402,
      "kl": 0.095947265625,
      "learning_rate": 9.042553191489361e-07,
      "loss": 0.0038,
      "reward": 1.6154414415359497,
      "reward_std": 0.14917904138565063,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6154414117336273,
      "step": 297
    },
    {
      "completion_length": 100.515625,
      "epoch": 0.19213410702772404,
      "grad_norm": 13.001978874206543,
      "kl": 0.216064453125,
      "learning_rate": 9.03932946486138e-07,
      "loss": 0.0086,
      "reward": 1.6394542455673218,
      "reward_std": 0.17497223615646362,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6394542753696442,
      "step": 298
    },
    {
      "completion_length": 104.65625,
      "epoch": 0.19277885235332043,
      "grad_norm": 16.04650115966797,
      "kl": 0.09130859375,
      "learning_rate": 9.036105738233397e-07,
      "loss": 0.0037,
      "reward": 1.4991700649261475,
      "reward_std": 0.24890705943107605,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5147949606180191,
      "step": 299
    },
    {
      "completion_length": 104.203125,
      "epoch": 0.19342359767891681,
      "grad_norm": 19.988811492919922,
      "kl": 0.095458984375,
      "learning_rate": 9.032882011605415e-07,
      "loss": 0.0038,
      "reward": 1.547214686870575,
      "reward_std": 0.11229012906551361,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5472146570682526,
      "step": 300
    },
    {
      "completion_length": 97.921875,
      "epoch": 0.19406834300451323,
      "grad_norm": 15.380250930786133,
      "kl": 0.0888671875,
      "learning_rate": 9.029658284977433e-07,
      "loss": 0.0035,
      "reward": 1.6846535801887512,
      "reward_std": 0.11254275217652321,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6846535801887512,
      "step": 301
    },
    {
      "completion_length": 101.359375,
      "epoch": 0.1947130883301096,
      "grad_norm": 11.956006050109863,
      "kl": 0.091796875,
      "learning_rate": 9.026434558349452e-07,
      "loss": 0.0037,
      "reward": 1.584118127822876,
      "reward_std": 0.12083565443754196,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5841181576251984,
      "step": 302
    },
    {
      "completion_length": 107.328125,
      "epoch": 0.195357833655706,
      "grad_norm": 8.037882804870605,
      "kl": 0.0869140625,
      "learning_rate": 9.023210831721469e-07,
      "loss": 0.0035,
      "reward": 1.5585249662399292,
      "reward_std": 0.11393819749355316,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.558524951338768,
      "step": 303
    },
    {
      "completion_length": 93.875,
      "epoch": 0.19600257898130238,
      "grad_norm": 13.471504211425781,
      "kl": 0.105712890625,
      "learning_rate": 9.019987105093487e-07,
      "loss": 0.0042,
      "reward": 1.5376924872398376,
      "reward_std": 0.18834134191274643,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5376925766468048,
      "step": 304
    },
    {
      "completion_length": 92.96875,
      "epoch": 0.19664732430689877,
      "grad_norm": 10.125955581665039,
      "kl": 0.114501953125,
      "learning_rate": 9.016763378465506e-07,
      "loss": 0.0046,
      "reward": 1.5640528798103333,
      "reward_std": 0.14177117496728897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.564052939414978,
      "step": 305
    },
    {
      "completion_length": 98.96875,
      "epoch": 0.19729206963249515,
      "grad_norm": 9.222174644470215,
      "kl": 0.1259765625,
      "learning_rate": 9.013539651837524e-07,
      "loss": 0.005,
      "reward": 1.4195857048034668,
      "reward_std": 0.13147933036088943,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4195857346057892,
      "step": 306
    },
    {
      "completion_length": 87.921875,
      "epoch": 0.19793681495809157,
      "grad_norm": 11.84753704071045,
      "kl": 0.11865234375,
      "learning_rate": 9.010315925209541e-07,
      "loss": 0.0048,
      "reward": 1.5052205920219421,
      "reward_std": 0.15341130271553993,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5052205324172974,
      "step": 307
    },
    {
      "completion_length": 88.09375,
      "epoch": 0.19858156028368795,
      "grad_norm": 9.092905044555664,
      "kl": 0.113525390625,
      "learning_rate": 9.00709219858156e-07,
      "loss": 0.0045,
      "reward": 1.6300784349441528,
      "reward_std": 0.11801314726471901,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6300784051418304,
      "step": 308
    },
    {
      "completion_length": 89.203125,
      "epoch": 0.19922630560928434,
      "grad_norm": 13.911439895629883,
      "kl": 0.114990234375,
      "learning_rate": 9.003868471953578e-07,
      "loss": 0.0046,
      "reward": 1.4316323399543762,
      "reward_std": 0.19944678246974945,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4316323399543762,
      "step": 309
    },
    {
      "completion_length": 95.453125,
      "epoch": 0.19987105093488072,
      "grad_norm": 11.394854545593262,
      "kl": 0.097412109375,
      "learning_rate": 9.000644745325596e-07,
      "loss": 0.0039,
      "reward": 1.53744238615036,
      "reward_std": 0.15609167516231537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5374423861503601,
      "step": 310
    },
    {
      "completion_length": 90.5625,
      "epoch": 0.2005157962604771,
      "grad_norm": 43.32953643798828,
      "kl": 0.095458984375,
      "learning_rate": 8.997421018697614e-07,
      "loss": 0.0038,
      "reward": 1.5357195138931274,
      "reward_std": 0.14336901158094406,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5357194542884827,
      "step": 311
    },
    {
      "completion_length": 93.515625,
      "epoch": 0.2011605415860735,
      "grad_norm": 11.800814628601074,
      "kl": 0.0869140625,
      "learning_rate": 8.994197292069632e-07,
      "loss": 0.0035,
      "reward": 1.5805315971374512,
      "reward_std": 0.14307568967342377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5805316269397736,
      "step": 312
    },
    {
      "completion_length": 90.515625,
      "epoch": 0.20180528691166988,
      "grad_norm": 7.8137922286987305,
      "kl": 0.092529296875,
      "learning_rate": 8.990973565441649e-07,
      "loss": 0.0037,
      "reward": 1.6621150970458984,
      "reward_std": 0.09617417305707932,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6621150374412537,
      "step": 313
    },
    {
      "completion_length": 94.890625,
      "epoch": 0.2024500322372663,
      "grad_norm": 19.35686683654785,
      "kl": 0.28369140625,
      "learning_rate": 8.987749838813669e-07,
      "loss": 0.0113,
      "reward": 1.4318525791168213,
      "reward_std": 0.13407088071107864,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4318525493144989,
      "step": 314
    },
    {
      "completion_length": 96.921875,
      "epoch": 0.20309477756286268,
      "grad_norm": 9.18880558013916,
      "kl": 0.118408203125,
      "learning_rate": 8.984526112185686e-07,
      "loss": 0.0047,
      "reward": 1.4458122253417969,
      "reward_std": 0.11026818305253983,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4458121508359909,
      "step": 315
    },
    {
      "completion_length": 98.46875,
      "epoch": 0.20373952288845906,
      "grad_norm": 10.588865280151367,
      "kl": 0.10791015625,
      "learning_rate": 8.981302385557704e-07,
      "loss": 0.0043,
      "reward": 1.527498185634613,
      "reward_std": 0.13546320796012878,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5274982452392578,
      "step": 316
    },
    {
      "completion_length": 92.125,
      "epoch": 0.20438426821405545,
      "grad_norm": 57.36320877075195,
      "kl": 0.08984375,
      "learning_rate": 8.978078658929722e-07,
      "loss": 0.0036,
      "reward": 1.3975678086280823,
      "reward_std": 0.12371647357940674,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3975677937269211,
      "step": 317
    },
    {
      "completion_length": 91.84375,
      "epoch": 0.20502901353965183,
      "grad_norm": 10.51960563659668,
      "kl": 0.097900390625,
      "learning_rate": 8.974854932301741e-07,
      "loss": 0.0039,
      "reward": 1.5480912327766418,
      "reward_std": 0.06759429350495338,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5480912178754807,
      "step": 318
    },
    {
      "completion_length": 85.546875,
      "epoch": 0.20567375886524822,
      "grad_norm": 11.455018997192383,
      "kl": 0.11865234375,
      "learning_rate": 8.971631205673758e-07,
      "loss": 0.0047,
      "reward": 1.5190264582633972,
      "reward_std": 0.12998723611235619,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5190264880657196,
      "step": 319
    },
    {
      "completion_length": 88.8125,
      "epoch": 0.2063185041908446,
      "grad_norm": 11.84350299835205,
      "kl": 0.1015625,
      "learning_rate": 8.968407479045777e-07,
      "loss": 0.0041,
      "reward": 1.4973270297050476,
      "reward_std": 0.12015613168478012,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.497326985001564,
      "step": 320
    },
    {
      "completion_length": 91.34375,
      "epoch": 0.20696324951644102,
      "grad_norm": 16.877960205078125,
      "kl": 0.19580078125,
      "learning_rate": 8.965183752417794e-07,
      "loss": 0.0079,
      "reward": 1.4202344417572021,
      "reward_std": 0.12599121034145355,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4202345311641693,
      "step": 321
    },
    {
      "completion_length": 84.59375,
      "epoch": 0.2076079948420374,
      "grad_norm": 23.965744018554688,
      "kl": 0.1162109375,
      "learning_rate": 8.961960025789813e-07,
      "loss": 0.0047,
      "reward": 1.578860580921173,
      "reward_std": 0.10854113101959229,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5788605213165283,
      "step": 322
    },
    {
      "completion_length": 85.796875,
      "epoch": 0.2082527401676338,
      "grad_norm": 17.646039962768555,
      "kl": 0.104736328125,
      "learning_rate": 8.95873629916183e-07,
      "loss": 0.0042,
      "reward": 1.4003408551216125,
      "reward_std": 0.17851673811674118,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.40034082531929016,
      "step": 323
    },
    {
      "completion_length": 86.5625,
      "epoch": 0.20889748549323017,
      "grad_norm": 32.0064697265625,
      "kl": 0.09375,
      "learning_rate": 8.955512572533849e-07,
      "loss": 0.0038,
      "reward": 1.579477608203888,
      "reward_std": 0.1306827887892723,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5794776082038879,
      "step": 324
    },
    {
      "completion_length": 83.5625,
      "epoch": 0.20954223081882656,
      "grad_norm": 11.898215293884277,
      "kl": 0.099853515625,
      "learning_rate": 8.952288845905866e-07,
      "loss": 0.004,
      "reward": 1.4137381315231323,
      "reward_std": 0.09641918167471886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41373810172080994,
      "step": 325
    },
    {
      "completion_length": 83.96875,
      "epoch": 0.21018697614442294,
      "grad_norm": 11.983473777770996,
      "kl": 0.126220703125,
      "learning_rate": 8.949065119277884e-07,
      "loss": 0.0051,
      "reward": 1.5423098802566528,
      "reward_std": 0.16920984536409378,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5579348504543304,
      "step": 326
    },
    {
      "completion_length": 86.546875,
      "epoch": 0.21083172147001933,
      "grad_norm": 10.262345314025879,
      "kl": 0.111328125,
      "learning_rate": 8.945841392649903e-07,
      "loss": 0.0045,
      "reward": 1.535430669784546,
      "reward_std": 0.1486879587173462,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5354305803775787,
      "step": 327
    },
    {
      "completion_length": 82.140625,
      "epoch": 0.21147646679561574,
      "grad_norm": 21.612335205078125,
      "kl": 0.10498046875,
      "learning_rate": 8.942617666021921e-07,
      "loss": 0.0042,
      "reward": 1.548956573009491,
      "reward_std": 0.09943357855081558,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.548956573009491,
      "step": 328
    },
    {
      "completion_length": 81.609375,
      "epoch": 0.21212121212121213,
      "grad_norm": 16.632413864135742,
      "kl": 0.115234375,
      "learning_rate": 8.939393939393938e-07,
      "loss": 0.0046,
      "reward": 1.5397294163703918,
      "reward_std": 0.16385426372289658,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5397294163703918,
      "step": 329
    },
    {
      "completion_length": 81.546875,
      "epoch": 0.2127659574468085,
      "grad_norm": 20.90030860900879,
      "kl": 0.13818359375,
      "learning_rate": 8.936170212765957e-07,
      "loss": 0.0055,
      "reward": 1.507832646369934,
      "reward_std": 0.16199873387813568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5078326165676117,
      "step": 330
    },
    {
      "completion_length": 88.640625,
      "epoch": 0.2134107027724049,
      "grad_norm": 13.776839256286621,
      "kl": 0.107421875,
      "learning_rate": 8.932946486137975e-07,
      "loss": 0.0043,
      "reward": 1.618999719619751,
      "reward_std": 0.1200912743806839,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6189996898174286,
      "step": 331
    },
    {
      "completion_length": 83.625,
      "epoch": 0.21405544809800128,
      "grad_norm": 43.87213134765625,
      "kl": 0.11083984375,
      "learning_rate": 8.929722759509993e-07,
      "loss": 0.0044,
      "reward": 1.6609066128730774,
      "reward_std": 0.152876615524292,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.660906583070755,
      "step": 332
    },
    {
      "completion_length": 92.03125,
      "epoch": 0.21470019342359767,
      "grad_norm": 13.790566444396973,
      "kl": 0.12109375,
      "learning_rate": 8.926499032882011e-07,
      "loss": 0.0048,
      "reward": 1.4482386112213135,
      "reward_std": 0.14547474682331085,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4482386112213135,
      "step": 333
    },
    {
      "completion_length": 84.859375,
      "epoch": 0.21534493874919408,
      "grad_norm": 14.70999813079834,
      "kl": 0.12548828125,
      "learning_rate": 8.923275306254029e-07,
      "loss": 0.005,
      "reward": 1.565877914428711,
      "reward_std": 0.1912159025669098,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5971279144287109,
      "step": 334
    },
    {
      "completion_length": 87.53125,
      "epoch": 0.21598968407479047,
      "grad_norm": 26.293325424194336,
      "kl": 0.115478515625,
      "learning_rate": 8.920051579626047e-07,
      "loss": 0.0046,
      "reward": 1.5629397630691528,
      "reward_std": 0.19451142847537994,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5785647630691528,
      "step": 335
    },
    {
      "completion_length": 87.296875,
      "epoch": 0.21663442940038685,
      "grad_norm": 20.95355796813965,
      "kl": 0.122314453125,
      "learning_rate": 8.916827852998066e-07,
      "loss": 0.0049,
      "reward": 1.6097609996795654,
      "reward_std": 0.14318327605724335,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6097610294818878,
      "step": 336
    },
    {
      "completion_length": 83.875,
      "epoch": 0.21727917472598324,
      "grad_norm": 19.546236038208008,
      "kl": 0.117431640625,
      "learning_rate": 8.913604126370083e-07,
      "loss": 0.0047,
      "reward": 1.6258500814437866,
      "reward_std": 0.09713875502347946,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6258500814437866,
      "step": 337
    },
    {
      "completion_length": 86.09375,
      "epoch": 0.21792392005157962,
      "grad_norm": 31.88779067993164,
      "kl": 0.108642578125,
      "learning_rate": 8.910380399742101e-07,
      "loss": 0.0043,
      "reward": 1.4809694290161133,
      "reward_std": 0.12653079256415367,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48096954822540283,
      "step": 338
    },
    {
      "completion_length": 88.796875,
      "epoch": 0.218568665377176,
      "grad_norm": 13.355061531066895,
      "kl": 0.12353515625,
      "learning_rate": 8.907156673114119e-07,
      "loss": 0.0049,
      "reward": 1.510122537612915,
      "reward_std": 0.180454570800066,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.525747537612915,
      "step": 339
    },
    {
      "completion_length": 88.578125,
      "epoch": 0.2192134107027724,
      "grad_norm": 12.257147789001465,
      "kl": 0.135009765625,
      "learning_rate": 8.903932946486138e-07,
      "loss": 0.0054,
      "reward": 1.6074494123458862,
      "reward_std": 0.1379043161869049,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6074493527412415,
      "step": 340
    },
    {
      "completion_length": 96.953125,
      "epoch": 0.2198581560283688,
      "grad_norm": 7.543994426727295,
      "kl": 0.11181640625,
      "learning_rate": 8.900709219858155e-07,
      "loss": 0.0045,
      "reward": 1.6192185282707214,
      "reward_std": 0.10273057222366333,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6192185282707214,
      "step": 341
    },
    {
      "completion_length": 93.640625,
      "epoch": 0.2205029013539652,
      "grad_norm": 16.90849494934082,
      "kl": 0.09912109375,
      "learning_rate": 8.897485493230174e-07,
      "loss": 0.004,
      "reward": 1.464130938053131,
      "reward_std": 0.185988649725914,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4641309380531311,
      "step": 342
    },
    {
      "completion_length": 87.296875,
      "epoch": 0.22114764667956158,
      "grad_norm": 10.599559783935547,
      "kl": 0.118896484375,
      "learning_rate": 8.894261766602191e-07,
      "loss": 0.0047,
      "reward": 1.5260052680969238,
      "reward_std": 0.14786972850561142,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5260052978992462,
      "step": 343
    },
    {
      "completion_length": 94.296875,
      "epoch": 0.22179239200515796,
      "grad_norm": 12.245553970336914,
      "kl": 0.11279296875,
      "learning_rate": 8.89103803997421e-07,
      "loss": 0.0045,
      "reward": 1.62826669216156,
      "reward_std": 0.11454515159130096,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6282666921615601,
      "step": 344
    },
    {
      "completion_length": 91.0,
      "epoch": 0.22243713733075435,
      "grad_norm": 16.552492141723633,
      "kl": 0.11572265625,
      "learning_rate": 8.887814313346229e-07,
      "loss": 0.0046,
      "reward": 1.4791878461837769,
      "reward_std": 0.13600347191095352,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47918783128261566,
      "step": 345
    },
    {
      "completion_length": 99.8125,
      "epoch": 0.22308188265635073,
      "grad_norm": 8.797539710998535,
      "kl": 0.1162109375,
      "learning_rate": 8.884590586718246e-07,
      "loss": 0.0046,
      "reward": 1.4839099049568176,
      "reward_std": 0.12259767949581146,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4839099049568176,
      "step": 346
    },
    {
      "completion_length": 94.953125,
      "epoch": 0.22372662798194712,
      "grad_norm": 20.40011978149414,
      "kl": 0.121337890625,
      "learning_rate": 8.881366860090263e-07,
      "loss": 0.0048,
      "reward": 1.41604483127594,
      "reward_std": 0.19619542360305786,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4160449057817459,
      "step": 347
    },
    {
      "completion_length": 99.390625,
      "epoch": 0.22437137330754353,
      "grad_norm": 10.037973403930664,
      "kl": 0.16357421875,
      "learning_rate": 8.878143133462282e-07,
      "loss": 0.0065,
      "reward": 1.6190119981765747,
      "reward_std": 0.10121146589517593,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6190120130777359,
      "step": 348
    },
    {
      "completion_length": 100.984375,
      "epoch": 0.22501611863313992,
      "grad_norm": 35.19863510131836,
      "kl": 0.1494140625,
      "learning_rate": 8.874919406834301e-07,
      "loss": 0.006,
      "reward": 1.537758231163025,
      "reward_std": 0.13214527815580368,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5377582460641861,
      "step": 349
    },
    {
      "completion_length": 110.046875,
      "epoch": 0.2256608639587363,
      "grad_norm": 42.346046447753906,
      "kl": 0.12353515625,
      "learning_rate": 8.871695680206318e-07,
      "loss": 0.0049,
      "reward": 1.5095826387405396,
      "reward_std": 0.11873700842261314,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5095826536417007,
      "step": 350
    },
    {
      "completion_length": 99.984375,
      "epoch": 0.2263056092843327,
      "grad_norm": 14.766871452331543,
      "kl": 0.117431640625,
      "learning_rate": 8.868471953578336e-07,
      "loss": 0.0047,
      "reward": 1.5179364085197449,
      "reward_std": 0.13833826035261154,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5179364234209061,
      "step": 351
    },
    {
      "completion_length": 111.25,
      "epoch": 0.22695035460992907,
      "grad_norm": 10.721341133117676,
      "kl": 0.117431640625,
      "learning_rate": 8.865248226950354e-07,
      "loss": 0.0047,
      "reward": 1.4805157780647278,
      "reward_std": 0.10306541621685028,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4805157780647278,
      "step": 352
    },
    {
      "completion_length": 116.0625,
      "epoch": 0.22759509993552546,
      "grad_norm": 20.140018463134766,
      "kl": 0.11474609375,
      "learning_rate": 8.862024500322373e-07,
      "loss": 0.0046,
      "reward": 1.5178296566009521,
      "reward_std": 0.1821778416633606,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5334546715021133,
      "step": 353
    },
    {
      "completion_length": 93.765625,
      "epoch": 0.22823984526112184,
      "grad_norm": 19.487287521362305,
      "kl": 0.13232421875,
      "learning_rate": 8.85880077369439e-07,
      "loss": 0.0053,
      "reward": 1.6278626918792725,
      "reward_std": 0.20768529176712036,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6434876918792725,
      "step": 354
    },
    {
      "completion_length": 99.484375,
      "epoch": 0.22888459058671826,
      "grad_norm": 10.036696434020996,
      "kl": 0.16455078125,
      "learning_rate": 8.855577047066409e-07,
      "loss": 0.0066,
      "reward": 1.4639315605163574,
      "reward_std": 0.11769643425941467,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4639315754175186,
      "step": 355
    },
    {
      "completion_length": 97.75,
      "epoch": 0.22952933591231464,
      "grad_norm": 9.284832954406738,
      "kl": 0.13671875,
      "learning_rate": 8.852353320438426e-07,
      "loss": 0.0055,
      "reward": 1.588053584098816,
      "reward_std": 0.17258089780807495,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5880535244941711,
      "step": 356
    },
    {
      "completion_length": 102.859375,
      "epoch": 0.23017408123791103,
      "grad_norm": 17.785030364990234,
      "kl": 0.158203125,
      "learning_rate": 8.849129593810445e-07,
      "loss": 0.0063,
      "reward": 1.5744596719741821,
      "reward_std": 0.11283202096819878,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5744596570730209,
      "step": 357
    },
    {
      "completion_length": 112.25,
      "epoch": 0.2308188265635074,
      "grad_norm": 8.444705963134766,
      "kl": 0.109375,
      "learning_rate": 8.845905867182463e-07,
      "loss": 0.0044,
      "reward": 1.6061638593673706,
      "reward_std": 0.15277552604675293,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6061639040708542,
      "step": 358
    },
    {
      "completion_length": 116.953125,
      "epoch": 0.2314635718891038,
      "grad_norm": 91.60777282714844,
      "kl": 0.1181640625,
      "learning_rate": 8.842682140554481e-07,
      "loss": 0.0047,
      "reward": 1.6188980340957642,
      "reward_std": 0.18260718137025833,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6345230638980865,
      "step": 359
    },
    {
      "completion_length": 110.375,
      "epoch": 0.23210831721470018,
      "grad_norm": 8.044037818908691,
      "kl": 0.129150390625,
      "learning_rate": 8.839458413926498e-07,
      "loss": 0.0052,
      "reward": 1.4037490487098694,
      "reward_std": 0.1555328145623207,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.40374905616045,
      "step": 360
    },
    {
      "completion_length": 109.671875,
      "epoch": 0.2327530625402966,
      "grad_norm": 10.870550155639648,
      "kl": 0.134033203125,
      "learning_rate": 8.836234687298518e-07,
      "loss": 0.0054,
      "reward": 1.46833074092865,
      "reward_std": 0.13861528784036636,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4683307707309723,
      "step": 361
    },
    {
      "completion_length": 100.71875,
      "epoch": 0.23339780786589298,
      "grad_norm": 22.924345016479492,
      "kl": 0.1484375,
      "learning_rate": 8.833010960670535e-07,
      "loss": 0.0059,
      "reward": 1.3982949256896973,
      "reward_std": 0.19406422972679138,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3982948958873749,
      "step": 362
    },
    {
      "completion_length": 115.578125,
      "epoch": 0.23404255319148937,
      "grad_norm": 10.494964599609375,
      "kl": 0.14404296875,
      "learning_rate": 8.829787234042553e-07,
      "loss": 0.0058,
      "reward": 1.4072777032852173,
      "reward_std": 0.126494862139225,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4229026734828949,
      "step": 363
    },
    {
      "completion_length": 108.265625,
      "epoch": 0.23468729851708575,
      "grad_norm": 21.195520401000977,
      "kl": 0.1591796875,
      "learning_rate": 8.82656350741457e-07,
      "loss": 0.0063,
      "reward": 1.6603394746780396,
      "reward_std": 0.18765471875667572,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6603395640850067,
      "step": 364
    },
    {
      "completion_length": 101.9375,
      "epoch": 0.23533204384268214,
      "grad_norm": 13.110071182250977,
      "kl": 0.13330078125,
      "learning_rate": 8.823339780786589e-07,
      "loss": 0.0053,
      "reward": 1.6690788865089417,
      "reward_std": 0.13631463050842285,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6690788567066193,
      "step": 365
    },
    {
      "completion_length": 120.15625,
      "epoch": 0.23597678916827852,
      "grad_norm": 10.89026165008545,
      "kl": 0.112060546875,
      "learning_rate": 8.820116054158607e-07,
      "loss": 0.0045,
      "reward": 1.3980477452278137,
      "reward_std": 0.19040542095899582,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39804771542549133,
      "step": 366
    },
    {
      "completion_length": 111.578125,
      "epoch": 0.2366215344938749,
      "grad_norm": 37.69853210449219,
      "kl": 0.144775390625,
      "learning_rate": 8.816892327530626e-07,
      "loss": 0.0058,
      "reward": 1.5169401168823242,
      "reward_std": 0.1048840656876564,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5169401168823242,
      "step": 367
    },
    {
      "completion_length": 127.703125,
      "epoch": 0.23726627981947132,
      "grad_norm": 8.726611137390137,
      "kl": 0.110107421875,
      "learning_rate": 8.813668600902643e-07,
      "loss": 0.0044,
      "reward": 1.6172541975975037,
      "reward_std": 0.15612297505140305,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6172541975975037,
      "step": 368
    },
    {
      "completion_length": 127.125,
      "epoch": 0.2379110251450677,
      "grad_norm": 9.400821685791016,
      "kl": 0.14892578125,
      "learning_rate": 8.810444874274661e-07,
      "loss": 0.006,
      "reward": 1.5110276341438293,
      "reward_std": 0.20710352808237076,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5266527086496353,
      "step": 369
    },
    {
      "completion_length": 116.203125,
      "epoch": 0.2385557704706641,
      "grad_norm": 10.621955871582031,
      "kl": 0.11181640625,
      "learning_rate": 8.807221147646679e-07,
      "loss": 0.0045,
      "reward": 1.5545547008514404,
      "reward_std": 0.1288330778479576,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5545547008514404,
      "step": 370
    },
    {
      "completion_length": 111.578125,
      "epoch": 0.23920051579626048,
      "grad_norm": 8.831900596618652,
      "kl": 0.122314453125,
      "learning_rate": 8.803997421018698e-07,
      "loss": 0.0049,
      "reward": 1.5220544338226318,
      "reward_std": 0.12251096218824387,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5220544338226318,
      "step": 371
    },
    {
      "completion_length": 122.109375,
      "epoch": 0.23984526112185686,
      "grad_norm": 25.54118537902832,
      "kl": 0.1123046875,
      "learning_rate": 8.800773694390715e-07,
      "loss": 0.0045,
      "reward": 1.5144230723381042,
      "reward_std": 0.17132049053907394,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.514423057436943,
      "step": 372
    },
    {
      "completion_length": 117.546875,
      "epoch": 0.24049000644745325,
      "grad_norm": 13.058128356933594,
      "kl": 0.146484375,
      "learning_rate": 8.797549967762733e-07,
      "loss": 0.0059,
      "reward": 1.5927224159240723,
      "reward_std": 0.11529594287276268,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5927222967147827,
      "step": 373
    },
    {
      "completion_length": 120.109375,
      "epoch": 0.24113475177304963,
      "grad_norm": 9.465697288513184,
      "kl": 0.1240234375,
      "learning_rate": 8.794326241134752e-07,
      "loss": 0.005,
      "reward": 1.5129834413528442,
      "reward_std": 0.14658623188734055,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.512983426451683,
      "step": 374
    },
    {
      "completion_length": 115.234375,
      "epoch": 0.24177949709864605,
      "grad_norm": 14.327066421508789,
      "kl": 0.116943359375,
      "learning_rate": 8.79110251450677e-07,
      "loss": 0.0047,
      "reward": 1.5295900106430054,
      "reward_std": 0.125333983451128,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5295900255441666,
      "step": 375
    },
    {
      "completion_length": 109.453125,
      "epoch": 0.24242424242424243,
      "grad_norm": 13.450315475463867,
      "kl": 0.1328125,
      "learning_rate": 8.787878787878787e-07,
      "loss": 0.0053,
      "reward": 1.5274213552474976,
      "reward_std": 0.1867852807044983,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5274212658405304,
      "step": 376
    },
    {
      "completion_length": 113.0,
      "epoch": 0.24306898774983882,
      "grad_norm": 38.73995590209961,
      "kl": 0.1494140625,
      "learning_rate": 8.784655061250806e-07,
      "loss": 0.006,
      "reward": 1.5954452753067017,
      "reward_std": 0.16821444034576416,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5954452753067017,
      "step": 377
    },
    {
      "completion_length": 113.75,
      "epoch": 0.2437137330754352,
      "grad_norm": 8.51379108428955,
      "kl": 0.1337890625,
      "learning_rate": 8.781431334622823e-07,
      "loss": 0.0054,
      "reward": 1.4244216084480286,
      "reward_std": 0.08954451978206635,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42442163825035095,
      "step": 378
    },
    {
      "completion_length": 118.125,
      "epoch": 0.2443584784010316,
      "grad_norm": 12.981401443481445,
      "kl": 0.121337890625,
      "learning_rate": 8.778207607994842e-07,
      "loss": 0.0049,
      "reward": 1.536249577999115,
      "reward_std": 0.20471828430891037,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.536249577999115,
      "step": 379
    },
    {
      "completion_length": 118.390625,
      "epoch": 0.24500322372662797,
      "grad_norm": 17.424579620361328,
      "kl": 0.1240234375,
      "learning_rate": 8.77498388136686e-07,
      "loss": 0.005,
      "reward": 1.4686583876609802,
      "reward_std": 0.12933894246816635,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4686584174633026,
      "step": 380
    },
    {
      "completion_length": 103.21875,
      "epoch": 0.24564796905222436,
      "grad_norm": 13.15453815460205,
      "kl": 0.1240234375,
      "learning_rate": 8.771760154738878e-07,
      "loss": 0.0049,
      "reward": 1.6250154376029968,
      "reward_std": 0.09519154578447342,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6250154674053192,
      "step": 381
    },
    {
      "completion_length": 116.5625,
      "epoch": 0.24629271437782077,
      "grad_norm": 10.82996654510498,
      "kl": 0.1396484375,
      "learning_rate": 8.768536428110895e-07,
      "loss": 0.0056,
      "reward": 1.631207287311554,
      "reward_std": 0.1291571594774723,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.631207287311554,
      "step": 382
    },
    {
      "completion_length": 111.671875,
      "epoch": 0.24693745970341716,
      "grad_norm": 13.207661628723145,
      "kl": 0.1298828125,
      "learning_rate": 8.765312701482915e-07,
      "loss": 0.0052,
      "reward": 1.374193549156189,
      "reward_std": 0.13640347868204117,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37419362366199493,
      "step": 383
    },
    {
      "completion_length": 109.421875,
      "epoch": 0.24758220502901354,
      "grad_norm": 11.31387996673584,
      "kl": 0.12451171875,
      "learning_rate": 8.762088974854932e-07,
      "loss": 0.005,
      "reward": 1.560090184211731,
      "reward_std": 0.21742689609527588,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5757152438163757,
      "step": 384
    },
    {
      "completion_length": 105.734375,
      "epoch": 0.24822695035460993,
      "grad_norm": 46.513221740722656,
      "kl": 0.124267578125,
      "learning_rate": 8.75886524822695e-07,
      "loss": 0.005,
      "reward": 1.525193452835083,
      "reward_std": 0.1469801962375641,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.525193452835083,
      "step": 385
    },
    {
      "completion_length": 110.765625,
      "epoch": 0.2488716956802063,
      "grad_norm": 15.489583015441895,
      "kl": 0.122314453125,
      "learning_rate": 8.755641521598968e-07,
      "loss": 0.0049,
      "reward": 1.5546384453773499,
      "reward_std": 0.10428916290402412,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5546384751796722,
      "step": 386
    },
    {
      "completion_length": 113.890625,
      "epoch": 0.2495164410058027,
      "grad_norm": 14.360328674316406,
      "kl": 0.272216796875,
      "learning_rate": 8.752417794970987e-07,
      "loss": 0.011,
      "reward": 1.5682534575462341,
      "reward_std": 0.15514999628067017,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5838784277439117,
      "step": 387
    },
    {
      "completion_length": 119.609375,
      "epoch": 0.2501611863313991,
      "grad_norm": 8.095773696899414,
      "kl": 0.109619140625,
      "learning_rate": 8.749194068343004e-07,
      "loss": 0.0044,
      "reward": 1.521271824836731,
      "reward_std": 0.10073830187320709,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5212718695402145,
      "step": 388
    },
    {
      "completion_length": 116.34375,
      "epoch": 0.2508059316569955,
      "grad_norm": 10.134056091308594,
      "kl": 0.11962890625,
      "learning_rate": 8.745970341715023e-07,
      "loss": 0.0048,
      "reward": 1.4365845918655396,
      "reward_std": 0.13728945702314377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43658460676670074,
      "step": 389
    },
    {
      "completion_length": 104.21875,
      "epoch": 0.2514506769825919,
      "grad_norm": 24.610681533813477,
      "kl": 0.1904296875,
      "learning_rate": 8.74274661508704e-07,
      "loss": 0.0076,
      "reward": 1.4568537473678589,
      "reward_std": 0.12321500480175018,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4724787175655365,
      "step": 390
    },
    {
      "completion_length": 114.578125,
      "epoch": 0.25209542230818827,
      "grad_norm": 12.346039772033691,
      "kl": 0.117431640625,
      "learning_rate": 8.739522888459059e-07,
      "loss": 0.0047,
      "reward": 1.5427250862121582,
      "reward_std": 0.11714871600270271,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5427251160144806,
      "step": 391
    },
    {
      "completion_length": 117.171875,
      "epoch": 0.25274016763378465,
      "grad_norm": 10.353236198425293,
      "kl": 0.109619140625,
      "learning_rate": 8.736299161831076e-07,
      "loss": 0.0044,
      "reward": 1.7059763073921204,
      "reward_std": 0.16882120072841644,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7216013371944427,
      "step": 392
    },
    {
      "completion_length": 117.4375,
      "epoch": 0.25338491295938104,
      "grad_norm": 28.252819061279297,
      "kl": 0.10791015625,
      "learning_rate": 8.733075435203095e-07,
      "loss": 0.0043,
      "reward": 1.5156517624855042,
      "reward_std": 0.1431019902229309,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5156517624855042,
      "step": 393
    },
    {
      "completion_length": 114.4375,
      "epoch": 0.2540296582849774,
      "grad_norm": 45.27177429199219,
      "kl": 0.117431640625,
      "learning_rate": 8.729851708575112e-07,
      "loss": 0.0047,
      "reward": 1.4806265234947205,
      "reward_std": 0.22300370782613754,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.5275015234947205,
      "step": 394
    },
    {
      "completion_length": 113.390625,
      "epoch": 0.2546744036105738,
      "grad_norm": 14.209746360778809,
      "kl": 0.12060546875,
      "learning_rate": 8.72662798194713e-07,
      "loss": 0.0048,
      "reward": 1.585417628288269,
      "reward_std": 0.14653317630290985,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6010426878929138,
      "step": 395
    },
    {
      "completion_length": 123.015625,
      "epoch": 0.2553191489361702,
      "grad_norm": 6.456853866577148,
      "kl": 0.114501953125,
      "learning_rate": 8.723404255319149e-07,
      "loss": 0.0046,
      "reward": 1.694661259651184,
      "reward_std": 0.17132965475320816,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7102862596511841,
      "step": 396
    },
    {
      "completion_length": 133.40625,
      "epoch": 0.2559638942617666,
      "grad_norm": 11.757306098937988,
      "kl": 0.12060546875,
      "learning_rate": 8.720180528691167e-07,
      "loss": 0.0048,
      "reward": 1.5562049746513367,
      "reward_std": 0.2403947338461876,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5874549746513367,
      "step": 397
    },
    {
      "completion_length": 124.859375,
      "epoch": 0.256608639587363,
      "grad_norm": 15.601590156555176,
      "kl": 0.1201171875,
      "learning_rate": 8.716956802063184e-07,
      "loss": 0.0048,
      "reward": 1.3463094234466553,
      "reward_std": 0.255843885242939,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.3931843936443329,
      "step": 398
    },
    {
      "completion_length": 124.671875,
      "epoch": 0.2572533849129594,
      "grad_norm": 8.446572303771973,
      "kl": 0.1142578125,
      "learning_rate": 8.713733075435203e-07,
      "loss": 0.0046,
      "reward": 1.5371050834655762,
      "reward_std": 0.1252804920077324,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5527301132678986,
      "step": 399
    },
    {
      "completion_length": 123.171875,
      "epoch": 0.2578981302385558,
      "grad_norm": 9.509588241577148,
      "kl": 0.142578125,
      "learning_rate": 8.710509348807221e-07,
      "loss": 0.0057,
      "reward": 1.6943880319595337,
      "reward_std": 0.09968248754739761,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6943880319595337,
      "step": 400
    },
    {
      "completion_length": 115.953125,
      "epoch": 0.2585428755641522,
      "grad_norm": 26.3836612701416,
      "kl": 0.1240234375,
      "learning_rate": 8.707285622179239e-07,
      "loss": 0.005,
      "reward": 1.5159879922866821,
      "reward_std": 0.16381509974598885,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5316129773855209,
      "step": 401
    },
    {
      "completion_length": 134.421875,
      "epoch": 0.25918762088974856,
      "grad_norm": 12.983834266662598,
      "kl": 0.105712890625,
      "learning_rate": 8.704061895551257e-07,
      "loss": 0.0042,
      "reward": 1.4239293932914734,
      "reward_std": 0.19441810250282288,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.4551793485879898,
      "step": 402
    },
    {
      "completion_length": 119.921875,
      "epoch": 0.25983236621534495,
      "grad_norm": 19.93628692626953,
      "kl": 0.114013671875,
      "learning_rate": 8.700838168923275e-07,
      "loss": 0.0046,
      "reward": 1.525950849056244,
      "reward_std": 0.1031199786812067,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5415758490562439,
      "step": 403
    },
    {
      "completion_length": 119.359375,
      "epoch": 0.26047711154094133,
      "grad_norm": 9.794746398925781,
      "kl": 0.115966796875,
      "learning_rate": 8.697614442295293e-07,
      "loss": 0.0046,
      "reward": 1.4721354842185974,
      "reward_std": 0.19477830827236176,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4877604842185974,
      "step": 404
    },
    {
      "completion_length": 128.75,
      "epoch": 0.2611218568665377,
      "grad_norm": 11.187701225280762,
      "kl": 0.105224609375,
      "learning_rate": 8.694390715667312e-07,
      "loss": 0.0042,
      "reward": 1.5111178755760193,
      "reward_std": 0.13802172988653183,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5111178755760193,
      "step": 405
    },
    {
      "completion_length": 130.578125,
      "epoch": 0.2617666021921341,
      "grad_norm": 11.691713333129883,
      "kl": 0.0966796875,
      "learning_rate": 8.691166989039329e-07,
      "loss": 0.0039,
      "reward": 1.3761667013168335,
      "reward_std": 0.1583014875650406,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.3917917311191559,
      "step": 406
    },
    {
      "completion_length": 128.6875,
      "epoch": 0.2624113475177305,
      "grad_norm": 10.686534881591797,
      "kl": 0.103759765625,
      "learning_rate": 8.687943262411347e-07,
      "loss": 0.0041,
      "reward": 1.5369993448257446,
      "reward_std": 0.09643547236919403,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5369993597269058,
      "step": 407
    },
    {
      "completion_length": 117.6875,
      "epoch": 0.26305609284332687,
      "grad_norm": 39.99586868286133,
      "kl": 0.10400390625,
      "learning_rate": 8.684719535783365e-07,
      "loss": 0.0042,
      "reward": 1.606239914894104,
      "reward_std": 0.12021593004465103,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.606239914894104,
      "step": 408
    },
    {
      "completion_length": 117.03125,
      "epoch": 0.26370083816892326,
      "grad_norm": 10.232660293579102,
      "kl": 0.111572265625,
      "learning_rate": 8.681495809155384e-07,
      "loss": 0.0045,
      "reward": 1.7151300311088562,
      "reward_std": 0.09283937886357307,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7151300609111786,
      "step": 409
    },
    {
      "completion_length": 128.359375,
      "epoch": 0.26434558349451964,
      "grad_norm": 5.87454080581665,
      "kl": 0.10498046875,
      "learning_rate": 8.678272082527401e-07,
      "loss": 0.0042,
      "reward": 1.7446652054786682,
      "reward_std": 0.1887577325105667,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7602902054786682,
      "step": 410
    },
    {
      "completion_length": 126.859375,
      "epoch": 0.26499032882011603,
      "grad_norm": 16.121383666992188,
      "kl": 0.1181640625,
      "learning_rate": 8.67504835589942e-07,
      "loss": 0.0047,
      "reward": 1.6039236783981323,
      "reward_std": 0.10203766822814941,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6039237380027771,
      "step": 411
    },
    {
      "completion_length": 119.734375,
      "epoch": 0.26563507414571247,
      "grad_norm": 17.275545120239258,
      "kl": 0.14306640625,
      "learning_rate": 8.671824629271437e-07,
      "loss": 0.0057,
      "reward": 1.4978615641593933,
      "reward_std": 0.2517547532916069,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5134866237640381,
      "step": 412
    },
    {
      "completion_length": 135.46875,
      "epoch": 0.26627981947130885,
      "grad_norm": 15.758459091186523,
      "kl": 0.111328125,
      "learning_rate": 8.668600902643456e-07,
      "loss": 0.0045,
      "reward": 1.525922417640686,
      "reward_std": 0.18179912865161896,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5415474027395248,
      "step": 413
    },
    {
      "completion_length": 120.140625,
      "epoch": 0.26692456479690524,
      "grad_norm": 23.685686111450195,
      "kl": 0.1123046875,
      "learning_rate": 8.665377176015473e-07,
      "loss": 0.0045,
      "reward": 1.5499253273010254,
      "reward_std": 0.20756025612354279,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.565550372004509,
      "step": 414
    },
    {
      "completion_length": 107.4375,
      "epoch": 0.2675693101225016,
      "grad_norm": 17.447803497314453,
      "kl": 0.1376953125,
      "learning_rate": 8.662153449387492e-07,
      "loss": 0.0055,
      "reward": 1.4381148219108582,
      "reward_std": 0.21692107617855072,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.46936482191085815,
      "step": 415
    },
    {
      "completion_length": 110.953125,
      "epoch": 0.268214055448098,
      "grad_norm": 22.098020553588867,
      "kl": 0.1083984375,
      "learning_rate": 8.658929722759509e-07,
      "loss": 0.0043,
      "reward": 1.5825045108795166,
      "reward_std": 0.15497270971536636,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5981295108795166,
      "step": 416
    },
    {
      "completion_length": 120.234375,
      "epoch": 0.2688588007736944,
      "grad_norm": 18.107759475708008,
      "kl": 0.109375,
      "learning_rate": 8.655705996131528e-07,
      "loss": 0.0044,
      "reward": 1.4336537718772888,
      "reward_std": 0.12037287652492523,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4336537718772888,
      "step": 417
    },
    {
      "completion_length": 119.78125,
      "epoch": 0.2695035460992908,
      "grad_norm": 16.09417724609375,
      "kl": 0.119140625,
      "learning_rate": 8.652482269503546e-07,
      "loss": 0.0048,
      "reward": 1.4209988117218018,
      "reward_std": 0.22898180782794952,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.45224879682064056,
      "step": 418
    },
    {
      "completion_length": 118.421875,
      "epoch": 0.27014829142488717,
      "grad_norm": 12.481720924377441,
      "kl": 0.110107421875,
      "learning_rate": 8.649258542875564e-07,
      "loss": 0.0044,
      "reward": 1.3904073238372803,
      "reward_std": 0.14730828255414963,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4060323089361191,
      "step": 419
    },
    {
      "completion_length": 116.46875,
      "epoch": 0.27079303675048355,
      "grad_norm": 15.177681922912598,
      "kl": 0.1171875,
      "learning_rate": 8.646034816247581e-07,
      "loss": 0.0047,
      "reward": 1.521640121936798,
      "reward_std": 0.14473527297377586,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5372650921344757,
      "step": 420
    },
    {
      "completion_length": 106.96875,
      "epoch": 0.27143778207607994,
      "grad_norm": 15.35623550415039,
      "kl": 0.107421875,
      "learning_rate": 8.6428110896196e-07,
      "loss": 0.0043,
      "reward": 1.4848104119300842,
      "reward_std": 0.17813602089881897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48481038212776184,
      "step": 421
    },
    {
      "completion_length": 115.203125,
      "epoch": 0.2720825274016763,
      "grad_norm": 8.478485107421875,
      "kl": 0.17919921875,
      "learning_rate": 8.639587362991618e-07,
      "loss": 0.0071,
      "reward": 1.492174744606018,
      "reward_std": 0.10588141530752182,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49217478930950165,
      "step": 422
    },
    {
      "completion_length": 116.78125,
      "epoch": 0.2727272727272727,
      "grad_norm": 20.47816276550293,
      "kl": 0.106201171875,
      "learning_rate": 8.636363636363636e-07,
      "loss": 0.0042,
      "reward": 1.6990843415260315,
      "reward_std": 0.08219916746020317,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6990842819213867,
      "step": 423
    },
    {
      "completion_length": 108.609375,
      "epoch": 0.2733720180528691,
      "grad_norm": 15.236698150634766,
      "kl": 0.102294921875,
      "learning_rate": 8.633139909735654e-07,
      "loss": 0.0041,
      "reward": 1.5470861196517944,
      "reward_std": 0.18373820185661316,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5470860600471497,
      "step": 424
    },
    {
      "completion_length": 116.09375,
      "epoch": 0.27401676337846553,
      "grad_norm": 41.5452880859375,
      "kl": 0.10888671875,
      "learning_rate": 8.629916183107672e-07,
      "loss": 0.0044,
      "reward": 1.6228560209274292,
      "reward_std": 0.15928027778863907,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6228560507297516,
      "step": 425
    },
    {
      "completion_length": 106.75,
      "epoch": 0.2746615087040619,
      "grad_norm": 11.794381141662598,
      "kl": 0.110107421875,
      "learning_rate": 8.62669245647969e-07,
      "loss": 0.0044,
      "reward": 1.5028586387634277,
      "reward_std": 0.12982606142759323,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5028586536645889,
      "step": 426
    },
    {
      "completion_length": 104.59375,
      "epoch": 0.2753062540296583,
      "grad_norm": 14.104724884033203,
      "kl": 0.107177734375,
      "learning_rate": 8.623468729851709e-07,
      "loss": 0.0043,
      "reward": 1.6256231665611267,
      "reward_std": 0.07264291495084763,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6256231665611267,
      "step": 427
    },
    {
      "completion_length": 95.453125,
      "epoch": 0.2759509993552547,
      "grad_norm": 12.817341804504395,
      "kl": 0.102783203125,
      "learning_rate": 8.620245003223726e-07,
      "loss": 0.0041,
      "reward": 1.4218062162399292,
      "reward_std": 0.21968971192836761,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4218061864376068,
      "step": 428
    },
    {
      "completion_length": 107.609375,
      "epoch": 0.2765957446808511,
      "grad_norm": 65.80889129638672,
      "kl": 0.10986328125,
      "learning_rate": 8.617021276595744e-07,
      "loss": 0.0044,
      "reward": 1.4983304142951965,
      "reward_std": 0.1709209829568863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49833041429519653,
      "step": 429
    },
    {
      "completion_length": 108.625,
      "epoch": 0.27724049000644746,
      "grad_norm": 9.083003044128418,
      "kl": 0.11669921875,
      "learning_rate": 8.613797549967763e-07,
      "loss": 0.0047,
      "reward": 1.3649240136146545,
      "reward_std": 0.09722113236784935,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.36492396891117096,
      "step": 430
    },
    {
      "completion_length": 108.3125,
      "epoch": 0.27788523533204385,
      "grad_norm": 13.82675838470459,
      "kl": 0.10009765625,
      "learning_rate": 8.610573823339781e-07,
      "loss": 0.004,
      "reward": 1.5768791437149048,
      "reward_std": 0.17914673686027527,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5925041735172272,
      "step": 431
    },
    {
      "completion_length": 112.28125,
      "epoch": 0.27852998065764023,
      "grad_norm": 8.703876495361328,
      "kl": 0.099365234375,
      "learning_rate": 8.607350096711798e-07,
      "loss": 0.004,
      "reward": 1.6096728444099426,
      "reward_std": 0.11242663115262985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6096728146076202,
      "step": 432
    },
    {
      "completion_length": 92.0,
      "epoch": 0.2791747259832366,
      "grad_norm": 21.1232852935791,
      "kl": 0.110107421875,
      "learning_rate": 8.604126370083817e-07,
      "loss": 0.0044,
      "reward": 1.66538667678833,
      "reward_std": 0.11511220782995224,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6653867065906525,
      "step": 433
    },
    {
      "completion_length": 114.4375,
      "epoch": 0.279819471308833,
      "grad_norm": 10.83615493774414,
      "kl": 0.13525390625,
      "learning_rate": 8.600902643455834e-07,
      "loss": 0.0054,
      "reward": 1.6429526805877686,
      "reward_std": 0.12804623320698738,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6429526805877686,
      "step": 434
    },
    {
      "completion_length": 93.515625,
      "epoch": 0.2804642166344294,
      "grad_norm": 286.1723327636719,
      "kl": 0.116943359375,
      "learning_rate": 8.597678916827853e-07,
      "loss": 0.0047,
      "reward": 1.4603025913238525,
      "reward_std": 0.12294166535139084,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4603026509284973,
      "step": 435
    },
    {
      "completion_length": 103.96875,
      "epoch": 0.28110896196002577,
      "grad_norm": 18.53604507446289,
      "kl": 0.1162109375,
      "learning_rate": 8.59445519019987e-07,
      "loss": 0.0047,
      "reward": 1.463842749595642,
      "reward_std": 0.13263465836644173,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4638426899909973,
      "step": 436
    },
    {
      "completion_length": 109.703125,
      "epoch": 0.28175370728562216,
      "grad_norm": 7.604578495025635,
      "kl": 0.0927734375,
      "learning_rate": 8.591231463571889e-07,
      "loss": 0.0037,
      "reward": 1.6767799854278564,
      "reward_std": 0.1197793260216713,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6767799556255341,
      "step": 437
    },
    {
      "completion_length": 110.15625,
      "epoch": 0.28239845261121854,
      "grad_norm": 12.251392364501953,
      "kl": 0.114013671875,
      "learning_rate": 8.588007736943906e-07,
      "loss": 0.0046,
      "reward": 1.6475961804389954,
      "reward_std": 0.16608282178640366,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.647596150636673,
      "step": 438
    },
    {
      "completion_length": 101.640625,
      "epoch": 0.283043197936815,
      "grad_norm": 45.992103576660156,
      "kl": 0.099609375,
      "learning_rate": 8.584784010315925e-07,
      "loss": 0.004,
      "reward": 1.530308485031128,
      "reward_std": 0.10350951924920082,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5303084552288055,
      "step": 439
    },
    {
      "completion_length": 112.5,
      "epoch": 0.28368794326241137,
      "grad_norm": 21.31104850769043,
      "kl": 0.102294921875,
      "learning_rate": 8.581560283687943e-07,
      "loss": 0.0041,
      "reward": 1.5717370510101318,
      "reward_std": 0.1433250457048416,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5717370808124542,
      "step": 440
    },
    {
      "completion_length": 116.625,
      "epoch": 0.28433268858800775,
      "grad_norm": 12.857828140258789,
      "kl": 0.1220703125,
      "learning_rate": 8.578336557059961e-07,
      "loss": 0.0049,
      "reward": 1.6379773020744324,
      "reward_std": 0.10524650290608406,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6379773616790771,
      "step": 441
    },
    {
      "completion_length": 120.796875,
      "epoch": 0.28497743391360414,
      "grad_norm": 14.907901763916016,
      "kl": 0.12646484375,
      "learning_rate": 8.575112830431978e-07,
      "loss": 0.005,
      "reward": 1.6316999793052673,
      "reward_std": 0.10639721155166626,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.631699949502945,
      "step": 442
    },
    {
      "completion_length": 110.0,
      "epoch": 0.2856221792392005,
      "grad_norm": 13.865013122558594,
      "kl": 0.1142578125,
      "learning_rate": 8.571889103803998e-07,
      "loss": 0.0046,
      "reward": 1.489943504333496,
      "reward_std": 0.08053775876760483,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48994359374046326,
      "step": 443
    },
    {
      "completion_length": 121.109375,
      "epoch": 0.2862669245647969,
      "grad_norm": 10.000602722167969,
      "kl": 0.103271484375,
      "learning_rate": 8.568665377176015e-07,
      "loss": 0.0041,
      "reward": 1.5654188394546509,
      "reward_std": 0.11091581359505653,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5654188394546509,
      "step": 444
    },
    {
      "completion_length": 121.9375,
      "epoch": 0.2869116698903933,
      "grad_norm": 15.2792329788208,
      "kl": 0.1064453125,
      "learning_rate": 8.565441650548033e-07,
      "loss": 0.0043,
      "reward": 1.6336507201194763,
      "reward_std": 0.10441675782203674,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6336507201194763,
      "step": 445
    },
    {
      "completion_length": 124.203125,
      "epoch": 0.2875564152159897,
      "grad_norm": 9.192334175109863,
      "kl": 0.1162109375,
      "learning_rate": 8.562217923920051e-07,
      "loss": 0.0046,
      "reward": 1.5857059359550476,
      "reward_std": 0.083042461425066,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5857059359550476,
      "step": 446
    },
    {
      "completion_length": 126.328125,
      "epoch": 0.28820116054158607,
      "grad_norm": 12.716821670532227,
      "kl": 0.0986328125,
      "learning_rate": 8.558994197292069e-07,
      "loss": 0.0039,
      "reward": 1.3910475969314575,
      "reward_std": 0.20336882025003433,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3910476565361023,
      "step": 447
    },
    {
      "completion_length": 98.984375,
      "epoch": 0.28884590586718245,
      "grad_norm": 19.093204498291016,
      "kl": 0.111328125,
      "learning_rate": 8.555770470664087e-07,
      "loss": 0.0044,
      "reward": 1.4914937615394592,
      "reward_std": 0.07705500349402428,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49149373173713684,
      "step": 448
    },
    {
      "completion_length": 117.171875,
      "epoch": 0.28949065119277884,
      "grad_norm": 10.103181838989258,
      "kl": 0.111328125,
      "learning_rate": 8.552546744036106e-07,
      "loss": 0.0045,
      "reward": 1.586841642856598,
      "reward_std": 0.15257032215595245,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5868416428565979,
      "step": 449
    },
    {
      "completion_length": 107.703125,
      "epoch": 0.2901353965183752,
      "grad_norm": 13.13296127319336,
      "kl": 0.111328125,
      "learning_rate": 8.549323017408123e-07,
      "loss": 0.0045,
      "reward": 1.52940034866333,
      "reward_std": 0.18080459535121918,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5294004082679749,
      "step": 450
    },
    {
      "completion_length": 110.34375,
      "epoch": 0.2907801418439716,
      "grad_norm": 12.052055358886719,
      "kl": 0.124755859375,
      "learning_rate": 8.546099290780141e-07,
      "loss": 0.005,
      "reward": 1.4507166743278503,
      "reward_std": 0.19025427103042603,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45071670413017273,
      "step": 451
    },
    {
      "completion_length": 104.078125,
      "epoch": 0.29142488716956805,
      "grad_norm": 12.829391479492188,
      "kl": 0.1220703125,
      "learning_rate": 8.54287556415216e-07,
      "loss": 0.0049,
      "reward": 1.6305707693099976,
      "reward_std": 0.18986354023218155,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.6618208289146423,
      "step": 452
    },
    {
      "completion_length": 103.703125,
      "epoch": 0.29206963249516443,
      "grad_norm": 8.33245849609375,
      "kl": 0.144775390625,
      "learning_rate": 8.539651837524178e-07,
      "loss": 0.0058,
      "reward": 1.596256971359253,
      "reward_std": 0.12559916824102402,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5962570309638977,
      "step": 453
    },
    {
      "completion_length": 95.34375,
      "epoch": 0.2927143778207608,
      "grad_norm": 19.02564239501953,
      "kl": 0.15234375,
      "learning_rate": 8.536428110896195e-07,
      "loss": 0.0061,
      "reward": 1.6366922855377197,
      "reward_std": 0.18177789449691772,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6366923153400421,
      "step": 454
    },
    {
      "completion_length": 107.3125,
      "epoch": 0.2933591231463572,
      "grad_norm": 109.59540557861328,
      "kl": 0.129150390625,
      "learning_rate": 8.533204384268214e-07,
      "loss": 0.0052,
      "reward": 1.5032501220703125,
      "reward_std": 0.14116708934307098,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5032500624656677,
      "step": 455
    },
    {
      "completion_length": 109.53125,
      "epoch": 0.2940038684719536,
      "grad_norm": 53.494632720947266,
      "kl": 0.11865234375,
      "learning_rate": 8.529980657640232e-07,
      "loss": 0.0047,
      "reward": 1.5563023090362549,
      "reward_std": 0.11787058785557747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5563022345304489,
      "step": 456
    },
    {
      "completion_length": 111.515625,
      "epoch": 0.29464861379755,
      "grad_norm": 23.822132110595703,
      "kl": 0.127197265625,
      "learning_rate": 8.52675693101225e-07,
      "loss": 0.0051,
      "reward": 1.4594323635101318,
      "reward_std": 0.05473834462463856,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45943234860897064,
      "step": 457
    },
    {
      "completion_length": 98.34375,
      "epoch": 0.29529335912314636,
      "grad_norm": 24.85257339477539,
      "kl": 0.13916015625,
      "learning_rate": 8.523533204384267e-07,
      "loss": 0.0056,
      "reward": 1.4711275696754456,
      "reward_std": 0.10964871942996979,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.471127524971962,
      "step": 458
    },
    {
      "completion_length": 105.453125,
      "epoch": 0.29593810444874274,
      "grad_norm": 14.3064546585083,
      "kl": 0.158203125,
      "learning_rate": 8.520309477756286e-07,
      "loss": 0.0063,
      "reward": 1.6756532788276672,
      "reward_std": 0.09594709798693657,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.675653338432312,
      "step": 459
    },
    {
      "completion_length": 105.96875,
      "epoch": 0.29658284977433913,
      "grad_norm": 32.8977165222168,
      "kl": 0.12890625,
      "learning_rate": 8.517085751128303e-07,
      "loss": 0.0052,
      "reward": 1.3314522504806519,
      "reward_std": 0.08916751854121685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33145226538181305,
      "step": 460
    },
    {
      "completion_length": 98.984375,
      "epoch": 0.2972275950999355,
      "grad_norm": 8.08194351196289,
      "kl": 0.125,
      "learning_rate": 8.513862024500322e-07,
      "loss": 0.005,
      "reward": 1.6798730492591858,
      "reward_std": 0.13914865255355835,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6798729598522186,
      "step": 461
    },
    {
      "completion_length": 104.1875,
      "epoch": 0.2978723404255319,
      "grad_norm": 17.793569564819336,
      "kl": 0.11328125,
      "learning_rate": 8.51063829787234e-07,
      "loss": 0.0045,
      "reward": 1.5144559741020203,
      "reward_std": 0.14659196138381958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5144559442996979,
      "step": 462
    },
    {
      "completion_length": 106.453125,
      "epoch": 0.2985170857511283,
      "grad_norm": 59.80522155761719,
      "kl": 0.171875,
      "learning_rate": 8.507414571244358e-07,
      "loss": 0.0069,
      "reward": 1.659844696521759,
      "reward_std": 0.1038866862654686,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6598446369171143,
      "step": 463
    },
    {
      "completion_length": 108.96875,
      "epoch": 0.29916183107672467,
      "grad_norm": 9.849380493164062,
      "kl": 0.11572265625,
      "learning_rate": 8.504190844616375e-07,
      "loss": 0.0046,
      "reward": 1.5677456259727478,
      "reward_std": 0.18492824956774712,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5833705961704254,
      "step": 464
    },
    {
      "completion_length": 102.890625,
      "epoch": 0.29980657640232106,
      "grad_norm": 9.406426429748535,
      "kl": 0.1396484375,
      "learning_rate": 8.500967117988395e-07,
      "loss": 0.0056,
      "reward": 1.5041118264198303,
      "reward_std": 0.13810846023261547,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5197368264198303,
      "step": 465
    },
    {
      "completion_length": 116.59375,
      "epoch": 0.3004513217279175,
      "grad_norm": 17.882394790649414,
      "kl": 0.126708984375,
      "learning_rate": 8.497743391360412e-07,
      "loss": 0.0051,
      "reward": 1.5210480093955994,
      "reward_std": 0.15733475238084793,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.521048054099083,
      "step": 466
    },
    {
      "completion_length": 102.96875,
      "epoch": 0.3010960670535139,
      "grad_norm": 17.0341854095459,
      "kl": 0.126220703125,
      "learning_rate": 8.49451966473243e-07,
      "loss": 0.0051,
      "reward": 1.420632779598236,
      "reward_std": 0.1144905835390091,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4362577497959137,
      "step": 467
    },
    {
      "completion_length": 94.65625,
      "epoch": 0.30174081237911027,
      "grad_norm": 14.085397720336914,
      "kl": 0.18115234375,
      "learning_rate": 8.491295938104448e-07,
      "loss": 0.0073,
      "reward": 1.5675341486930847,
      "reward_std": 0.19463778287172318,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5987841784954071,
      "step": 468
    },
    {
      "completion_length": 106.734375,
      "epoch": 0.30238555770470665,
      "grad_norm": 10.348201751708984,
      "kl": 0.136962890625,
      "learning_rate": 8.488072211476467e-07,
      "loss": 0.0055,
      "reward": 1.656339168548584,
      "reward_std": 0.2050868570804596,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6719641387462616,
      "step": 469
    },
    {
      "completion_length": 106.46875,
      "epoch": 0.30303030303030304,
      "grad_norm": 46.44258499145508,
      "kl": 0.123779296875,
      "learning_rate": 8.484848484848484e-07,
      "loss": 0.0049,
      "reward": 1.5097472667694092,
      "reward_std": 0.0910700261592865,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5097472071647644,
      "step": 470
    },
    {
      "completion_length": 99.203125,
      "epoch": 0.3036750483558994,
      "grad_norm": 25.939085006713867,
      "kl": 0.121337890625,
      "learning_rate": 8.481624758220503e-07,
      "loss": 0.0049,
      "reward": 1.5171257257461548,
      "reward_std": 0.14139147475361824,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.517125815153122,
      "step": 471
    },
    {
      "completion_length": 101.703125,
      "epoch": 0.3043197936814958,
      "grad_norm": 18.568218231201172,
      "kl": 0.116943359375,
      "learning_rate": 8.47840103159252e-07,
      "loss": 0.0047,
      "reward": 1.6284334659576416,
      "reward_std": 0.1252906247973442,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6284334659576416,
      "step": 472
    },
    {
      "completion_length": 112.78125,
      "epoch": 0.3049645390070922,
      "grad_norm": 21.57918357849121,
      "kl": 0.14306640625,
      "learning_rate": 8.475177304964538e-07,
      "loss": 0.0057,
      "reward": 1.4556211829185486,
      "reward_std": 0.08510977774858475,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4556211531162262,
      "step": 473
    },
    {
      "completion_length": 101.3125,
      "epoch": 0.3056092843326886,
      "grad_norm": 9.105512619018555,
      "kl": 0.125732421875,
      "learning_rate": 8.471953578336557e-07,
      "loss": 0.005,
      "reward": 1.4934914708137512,
      "reward_std": 0.08908221870660782,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4934914708137512,
      "step": 474
    },
    {
      "completion_length": 101.609375,
      "epoch": 0.30625402965828497,
      "grad_norm": 88.50373840332031,
      "kl": 0.1494140625,
      "learning_rate": 8.468729851708575e-07,
      "loss": 0.006,
      "reward": 1.3966038823127747,
      "reward_std": 0.14134737104177475,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39660392701625824,
      "step": 475
    },
    {
      "completion_length": 98.15625,
      "epoch": 0.30689877498388135,
      "grad_norm": 13.56151294708252,
      "kl": 0.13232421875,
      "learning_rate": 8.465506125080592e-07,
      "loss": 0.0053,
      "reward": 1.652710199356079,
      "reward_std": 0.12360907346010208,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6527101993560791,
      "step": 476
    },
    {
      "completion_length": 92.921875,
      "epoch": 0.30754352030947774,
      "grad_norm": 12.644444465637207,
      "kl": 0.14453125,
      "learning_rate": 8.46228239845261e-07,
      "loss": 0.0058,
      "reward": 1.5489562153816223,
      "reward_std": 0.10549729317426682,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5489561706781387,
      "step": 477
    },
    {
      "completion_length": 89.375,
      "epoch": 0.3081882656350741,
      "grad_norm": 35.62141036987305,
      "kl": 0.1298828125,
      "learning_rate": 8.459058671824629e-07,
      "loss": 0.0052,
      "reward": 1.4552032351493835,
      "reward_std": 0.1214294545352459,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45520325005054474,
      "step": 478
    },
    {
      "completion_length": 97.40625,
      "epoch": 0.30883301096067056,
      "grad_norm": 11.729783058166504,
      "kl": 0.13671875,
      "learning_rate": 8.455834945196647e-07,
      "loss": 0.0055,
      "reward": 1.6263262629508972,
      "reward_std": 0.16124407947063446,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6419512629508972,
      "step": 479
    },
    {
      "completion_length": 98.84375,
      "epoch": 0.30947775628626695,
      "grad_norm": 8.603642463684082,
      "kl": 0.14453125,
      "learning_rate": 8.452611218568664e-07,
      "loss": 0.0058,
      "reward": 1.4429956674575806,
      "reward_std": 0.15181055292487144,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44299568235874176,
      "step": 480
    },
    {
      "completion_length": 98.375,
      "epoch": 0.31012250161186333,
      "grad_norm": 19.082008361816406,
      "kl": 0.1376953125,
      "learning_rate": 8.449387491940683e-07,
      "loss": 0.0055,
      "reward": 1.5683542490005493,
      "reward_std": 0.1690295785665512,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5683543235063553,
      "step": 481
    },
    {
      "completion_length": 92.046875,
      "epoch": 0.3107672469374597,
      "grad_norm": 13.106690406799316,
      "kl": 0.1396484375,
      "learning_rate": 8.446163765312701e-07,
      "loss": 0.0056,
      "reward": 1.637206792831421,
      "reward_std": 0.11582134664058685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6372068524360657,
      "step": 482
    },
    {
      "completion_length": 94.5625,
      "epoch": 0.3114119922630561,
      "grad_norm": 14.102886199951172,
      "kl": 0.16455078125,
      "learning_rate": 8.44294003868472e-07,
      "loss": 0.0066,
      "reward": 1.555085003376007,
      "reward_std": 0.14258688688278198,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5707099884748459,
      "step": 483
    },
    {
      "completion_length": 97.15625,
      "epoch": 0.3120567375886525,
      "grad_norm": 15.093389511108398,
      "kl": 0.1337890625,
      "learning_rate": 8.439716312056737e-07,
      "loss": 0.0054,
      "reward": 1.6329968571662903,
      "reward_std": 0.13173328340053558,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6329968869686127,
      "step": 484
    },
    {
      "completion_length": 90.921875,
      "epoch": 0.3127014829142489,
      "grad_norm": 8.526525497436523,
      "kl": 0.185546875,
      "learning_rate": 8.436492585428755e-07,
      "loss": 0.0074,
      "reward": 1.5760527849197388,
      "reward_std": 0.12339088693261147,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5760527849197388,
      "step": 485
    },
    {
      "completion_length": 103.671875,
      "epoch": 0.31334622823984526,
      "grad_norm": 11.255599975585938,
      "kl": 0.16845703125,
      "learning_rate": 8.433268858800772e-07,
      "loss": 0.0068,
      "reward": 1.6047207713127136,
      "reward_std": 0.1114160493016243,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6047207415103912,
      "step": 486
    },
    {
      "completion_length": 108.015625,
      "epoch": 0.31399097356544164,
      "grad_norm": 18.66996955871582,
      "kl": 0.12158203125,
      "learning_rate": 8.430045132172792e-07,
      "loss": 0.0049,
      "reward": 1.6641413569450378,
      "reward_std": 0.0755637101829052,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6641413271427155,
      "step": 487
    },
    {
      "completion_length": 106.1875,
      "epoch": 0.31463571889103803,
      "grad_norm": 17.2938232421875,
      "kl": 0.12646484375,
      "learning_rate": 8.426821405544809e-07,
      "loss": 0.005,
      "reward": 1.3134232759475708,
      "reward_std": 0.09148961305618286,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.313423290848732,
      "step": 488
    },
    {
      "completion_length": 103.21875,
      "epoch": 0.3152804642166344,
      "grad_norm": 14.192865371704102,
      "kl": 0.16552734375,
      "learning_rate": 8.423597678916827e-07,
      "loss": 0.0066,
      "reward": 1.5219502449035645,
      "reward_std": 0.13911869004368782,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5375751405954361,
      "step": 489
    },
    {
      "completion_length": 104.734375,
      "epoch": 0.3159252095422308,
      "grad_norm": 8.525148391723633,
      "kl": 0.13623046875,
      "learning_rate": 8.420373952288845e-07,
      "loss": 0.0054,
      "reward": 1.4910805821418762,
      "reward_std": 0.14263714104890823,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4910805821418762,
      "step": 490
    },
    {
      "completion_length": 104.96875,
      "epoch": 0.3165699548678272,
      "grad_norm": 15.304465293884277,
      "kl": 0.16943359375,
      "learning_rate": 8.417150225660864e-07,
      "loss": 0.0068,
      "reward": 1.374679982662201,
      "reward_std": 0.0994303934276104,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37467992305755615,
      "step": 491
    },
    {
      "completion_length": 102.375,
      "epoch": 0.31721470019342357,
      "grad_norm": 28.216218948364258,
      "kl": 0.173828125,
      "learning_rate": 8.413926499032881e-07,
      "loss": 0.0069,
      "reward": 1.5741727948188782,
      "reward_std": 0.06946994923055172,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5741727650165558,
      "step": 492
    },
    {
      "completion_length": 101.3125,
      "epoch": 0.31785944551902,
      "grad_norm": 62.535240173339844,
      "kl": 0.15380859375,
      "learning_rate": 8.4107027724049e-07,
      "loss": 0.0061,
      "reward": 1.570895254611969,
      "reward_std": 0.16077566146850586,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5708952844142914,
      "step": 493
    },
    {
      "completion_length": 98.5625,
      "epoch": 0.3185041908446164,
      "grad_norm": 10.88364028930664,
      "kl": 0.162109375,
      "learning_rate": 8.407479045776917e-07,
      "loss": 0.0065,
      "reward": 1.5242938995361328,
      "reward_std": 0.12571700662374496,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.524293839931488,
      "step": 494
    },
    {
      "completion_length": 115.046875,
      "epoch": 0.3191489361702128,
      "grad_norm": 13.604912757873535,
      "kl": 0.140625,
      "learning_rate": 8.404255319148936e-07,
      "loss": 0.0056,
      "reward": 1.3118263483047485,
      "reward_std": 0.17000606656074524,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.32745136320590973,
      "step": 495
    },
    {
      "completion_length": 105.34375,
      "epoch": 0.31979368149580917,
      "grad_norm": 12.397253036499023,
      "kl": 0.1357421875,
      "learning_rate": 8.401031592520954e-07,
      "loss": 0.0054,
      "reward": 1.6170586347579956,
      "reward_std": 0.11738874390721321,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6170586943626404,
      "step": 496
    },
    {
      "completion_length": 100.828125,
      "epoch": 0.32043842682140555,
      "grad_norm": 34.156211853027344,
      "kl": 0.146484375,
      "learning_rate": 8.397807865892972e-07,
      "loss": 0.0058,
      "reward": 1.6565853357315063,
      "reward_std": 0.1069246418774128,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6565854251384735,
      "step": 497
    },
    {
      "completion_length": 103.625,
      "epoch": 0.32108317214700194,
      "grad_norm": 22.407413482666016,
      "kl": 0.1533203125,
      "learning_rate": 8.394584139264989e-07,
      "loss": 0.0061,
      "reward": 1.5076429843902588,
      "reward_std": 0.19160231947898865,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.52326799929142,
      "step": 498
    },
    {
      "completion_length": 100.53125,
      "epoch": 0.3217279174725983,
      "grad_norm": 12.296113967895508,
      "kl": 0.17041015625,
      "learning_rate": 8.391360412637009e-07,
      "loss": 0.0068,
      "reward": 1.560167908668518,
      "reward_std": 0.09707137942314148,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5601678788661957,
      "step": 499
    },
    {
      "completion_length": 109.15625,
      "epoch": 0.3223726627981947,
      "grad_norm": 8.46380615234375,
      "kl": 0.16357421875,
      "learning_rate": 8.388136686009026e-07,
      "loss": 0.0065,
      "reward": 1.460886836051941,
      "reward_std": 0.06691646948456764,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4608868658542633,
      "step": 500
    },
    {
      "completion_length": 103.703125,
      "epoch": 0.3230174081237911,
      "grad_norm": 57.8582763671875,
      "kl": 0.1689453125,
      "learning_rate": 8.384912959381044e-07,
      "loss": 0.0068,
      "reward": 1.6686757802963257,
      "reward_std": 0.12994926422834396,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6686758100986481,
      "step": 501
    },
    {
      "completion_length": 106.484375,
      "epoch": 0.3236621534493875,
      "grad_norm": 40.78071212768555,
      "kl": 0.14990234375,
      "learning_rate": 8.381689232753061e-07,
      "loss": 0.006,
      "reward": 1.4987604022026062,
      "reward_std": 0.11321312561631203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4987604022026062,
      "step": 502
    },
    {
      "completion_length": 112.4375,
      "epoch": 0.32430689877498386,
      "grad_norm": 14.115378379821777,
      "kl": 0.181640625,
      "learning_rate": 8.37846550612508e-07,
      "loss": 0.0073,
      "reward": 1.4107964038848877,
      "reward_std": 0.14325329661369324,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4107963293790817,
      "step": 503
    },
    {
      "completion_length": 95.15625,
      "epoch": 0.32495164410058025,
      "grad_norm": 12.940362930297852,
      "kl": 0.1630859375,
      "learning_rate": 8.375241779497098e-07,
      "loss": 0.0065,
      "reward": 1.4768180847167969,
      "reward_std": 0.17815791070461273,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4924430698156357,
      "step": 504
    },
    {
      "completion_length": 97.046875,
      "epoch": 0.32559638942617664,
      "grad_norm": 24.559680938720703,
      "kl": 0.173828125,
      "learning_rate": 8.372018052869116e-07,
      "loss": 0.007,
      "reward": 1.6828228831291199,
      "reward_std": 0.12030645087361336,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6828228235244751,
      "step": 505
    },
    {
      "completion_length": 103.046875,
      "epoch": 0.3262411347517731,
      "grad_norm": 9.820699691772461,
      "kl": 0.16650390625,
      "learning_rate": 8.368794326241134e-07,
      "loss": 0.0067,
      "reward": 1.6893452405929565,
      "reward_std": 0.057600412517786026,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6893452405929565,
      "step": 506
    },
    {
      "completion_length": 108.46875,
      "epoch": 0.32688588007736946,
      "grad_norm": 31.632915496826172,
      "kl": 0.1943359375,
      "learning_rate": 8.365570599613152e-07,
      "loss": 0.0078,
      "reward": 1.5965908765792847,
      "reward_std": 0.11272242292761803,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5965908467769623,
      "step": 507
    },
    {
      "completion_length": 92.796875,
      "epoch": 0.32753062540296585,
      "grad_norm": 13.336411476135254,
      "kl": 0.1728515625,
      "learning_rate": 8.36234687298517e-07,
      "loss": 0.0069,
      "reward": 1.5450761318206787,
      "reward_std": 0.08154022693634033,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5450761616230011,
      "step": 508
    },
    {
      "completion_length": 103.265625,
      "epoch": 0.32817537072856223,
      "grad_norm": 25.839597702026367,
      "kl": 0.1689453125,
      "learning_rate": 8.359123146357189e-07,
      "loss": 0.0067,
      "reward": 1.5176383256912231,
      "reward_std": 0.17332381755113602,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.517638310790062,
      "step": 509
    },
    {
      "completion_length": 95.1875,
      "epoch": 0.3288201160541586,
      "grad_norm": 12.781591415405273,
      "kl": 0.19091796875,
      "learning_rate": 8.355899419729206e-07,
      "loss": 0.0076,
      "reward": 1.3962189555168152,
      "reward_std": 0.07649979926645756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39621901512145996,
      "step": 510
    },
    {
      "completion_length": 100.734375,
      "epoch": 0.329464861379755,
      "grad_norm": 7.244078159332275,
      "kl": 0.17138671875,
      "learning_rate": 8.352675693101224e-07,
      "loss": 0.0069,
      "reward": 1.4728859066963196,
      "reward_std": 0.1124839149415493,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4885109066963196,
      "step": 511
    },
    {
      "completion_length": 100.515625,
      "epoch": 0.3301096067053514,
      "grad_norm": 22.949989318847656,
      "kl": 0.14453125,
      "learning_rate": 8.349451966473244e-07,
      "loss": 0.0058,
      "reward": 1.4569940567016602,
      "reward_std": 0.07238898053765297,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4569939970970154,
      "step": 512
    },
    {
      "completion_length": 96.140625,
      "epoch": 0.3307543520309478,
      "grad_norm": 11.172731399536133,
      "kl": 0.21484375,
      "learning_rate": 8.346228239845261e-07,
      "loss": 0.0086,
      "reward": 1.511184811592102,
      "reward_std": 0.12513361498713493,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5111847519874573,
      "step": 513
    },
    {
      "completion_length": 100.078125,
      "epoch": 0.33139909735654416,
      "grad_norm": 12.96349048614502,
      "kl": 0.1220703125,
      "learning_rate": 8.343004513217278e-07,
      "loss": 0.0049,
      "reward": 1.473563313484192,
      "reward_std": 0.1395437940955162,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4735632687807083,
      "step": 514
    },
    {
      "completion_length": 88.875,
      "epoch": 0.33204384268214054,
      "grad_norm": 12.410280227661133,
      "kl": 0.13525390625,
      "learning_rate": 8.339780786589297e-07,
      "loss": 0.0054,
      "reward": 1.6107569932937622,
      "reward_std": 0.13490690663456917,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6107569634914398,
      "step": 515
    },
    {
      "completion_length": 91.40625,
      "epoch": 0.33268858800773693,
      "grad_norm": 61.04192352294922,
      "kl": 0.1943359375,
      "learning_rate": 8.336557059961314e-07,
      "loss": 0.0078,
      "reward": 1.4572083353996277,
      "reward_std": 0.08159304037690163,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4572083503007889,
      "step": 516
    },
    {
      "completion_length": 108.03125,
      "epoch": 0.3333333333333333,
      "grad_norm": 18.190624237060547,
      "kl": 0.12451171875,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.005,
      "reward": 1.6369551420211792,
      "reward_std": 0.16067005693912506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6369551718235016,
      "step": 517
    },
    {
      "completion_length": 105.765625,
      "epoch": 0.3339780786589297,
      "grad_norm": 19.701316833496094,
      "kl": 0.12353515625,
      "learning_rate": 8.330109606705352e-07,
      "loss": 0.005,
      "reward": 1.5677099823951721,
      "reward_std": 0.13132942840456963,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5833349823951721,
      "step": 518
    },
    {
      "completion_length": 107.5,
      "epoch": 0.3346228239845261,
      "grad_norm": 11.90446949005127,
      "kl": 0.117431640625,
      "learning_rate": 8.326885880077369e-07,
      "loss": 0.0047,
      "reward": 1.6299498081207275,
      "reward_std": 0.10089965909719467,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6299497783184052,
      "step": 519
    },
    {
      "completion_length": 101.84375,
      "epoch": 0.3352675693101225,
      "grad_norm": 18.77952766418457,
      "kl": 0.111328125,
      "learning_rate": 8.323662153449386e-07,
      "loss": 0.0045,
      "reward": 1.4982839822769165,
      "reward_std": 0.0839691199362278,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4982839822769165,
      "step": 520
    },
    {
      "completion_length": 105.78125,
      "epoch": 0.3359123146357189,
      "grad_norm": 12.040685653686523,
      "kl": 0.102294921875,
      "learning_rate": 8.320438426821406e-07,
      "loss": 0.0041,
      "reward": 1.4064577221870422,
      "reward_std": 0.10008003562688828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4064577519893646,
      "step": 521
    },
    {
      "completion_length": 100.203125,
      "epoch": 0.3365570599613153,
      "grad_norm": 6.722482204437256,
      "kl": 0.13037109375,
      "learning_rate": 8.317214700193424e-07,
      "loss": 0.0052,
      "reward": 1.6532670259475708,
      "reward_std": 0.07082688249647617,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6532670557498932,
      "step": 522
    },
    {
      "completion_length": 102.390625,
      "epoch": 0.3372018052869117,
      "grad_norm": 20.486141204833984,
      "kl": 0.11865234375,
      "learning_rate": 8.313990973565441e-07,
      "loss": 0.0047,
      "reward": 1.6709571480751038,
      "reward_std": 0.1421183981001377,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6865821480751038,
      "step": 523
    },
    {
      "completion_length": 102.953125,
      "epoch": 0.33784655061250807,
      "grad_norm": 19.00589942932129,
      "kl": 0.13720703125,
      "learning_rate": 8.31076724693746e-07,
      "loss": 0.0055,
      "reward": 1.3803046345710754,
      "reward_std": 0.13267886638641357,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.395929679274559,
      "step": 524
    },
    {
      "completion_length": 121.15625,
      "epoch": 0.33849129593810445,
      "grad_norm": 28.536258697509766,
      "kl": 0.1142578125,
      "learning_rate": 8.307543520309478e-07,
      "loss": 0.0046,
      "reward": 1.4973403811454773,
      "reward_std": 0.13099077716469765,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4973403811454773,
      "step": 525
    },
    {
      "completion_length": 111.953125,
      "epoch": 0.33913604126370084,
      "grad_norm": 10.45938491821289,
      "kl": 0.09765625,
      "learning_rate": 8.304319793681496e-07,
      "loss": 0.0039,
      "reward": 1.5662608742713928,
      "reward_std": 0.13913103193044662,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5818859040737152,
      "step": 526
    },
    {
      "completion_length": 113.4375,
      "epoch": 0.3397807865892972,
      "grad_norm": 13.307291030883789,
      "kl": 0.24853515625,
      "learning_rate": 8.301096067053513e-07,
      "loss": 0.0099,
      "reward": 1.590022087097168,
      "reward_std": 0.18774433434009552,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.605647087097168,
      "step": 527
    },
    {
      "completion_length": 117.71875,
      "epoch": 0.3404255319148936,
      "grad_norm": 31.09686851501465,
      "kl": 0.101806640625,
      "learning_rate": 8.297872340425532e-07,
      "loss": 0.0041,
      "reward": 1.625194489955902,
      "reward_std": 0.14474449306726456,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6251944303512573,
      "step": 528
    },
    {
      "completion_length": 112.15625,
      "epoch": 0.34107027724049,
      "grad_norm": 13.974041938781738,
      "kl": 0.12744140625,
      "learning_rate": 8.294648613797549e-07,
      "loss": 0.0051,
      "reward": 1.5699716806411743,
      "reward_std": 0.1584780141711235,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5699716806411743,
      "step": 529
    },
    {
      "completion_length": 119.21875,
      "epoch": 0.3417150225660864,
      "grad_norm": 15.859521865844727,
      "kl": 0.102783203125,
      "learning_rate": 8.291424887169568e-07,
      "loss": 0.0041,
      "reward": 1.6906739473342896,
      "reward_std": 0.132080078125,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6906739473342896,
      "step": 530
    },
    {
      "completion_length": 124.0,
      "epoch": 0.34235976789168276,
      "grad_norm": 18.77686882019043,
      "kl": 0.1025390625,
      "learning_rate": 8.288201160541586e-07,
      "loss": 0.0041,
      "reward": 1.6462790966033936,
      "reward_std": 0.09448760747909546,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6462791562080383,
      "step": 531
    },
    {
      "completion_length": 124.125,
      "epoch": 0.34300451321727915,
      "grad_norm": 11.553679466247559,
      "kl": 0.11083984375,
      "learning_rate": 8.284977433913604e-07,
      "loss": 0.0044,
      "reward": 1.4024157524108887,
      "reward_std": 0.14840160310268402,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4024156928062439,
      "step": 532
    },
    {
      "completion_length": 118.765625,
      "epoch": 0.3436492585428756,
      "grad_norm": 39.30613708496094,
      "kl": 0.102783203125,
      "learning_rate": 8.281753707285621e-07,
      "loss": 0.0041,
      "reward": 1.5651786923408508,
      "reward_std": 0.14602860994637012,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.565178707242012,
      "step": 533
    },
    {
      "completion_length": 131.0,
      "epoch": 0.344294003868472,
      "grad_norm": 10.37143325805664,
      "kl": 0.119384765625,
      "learning_rate": 8.278529980657641e-07,
      "loss": 0.0048,
      "reward": 1.6771512627601624,
      "reward_std": 0.12305908650159836,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6771512627601624,
      "step": 534
    },
    {
      "completion_length": 106.59375,
      "epoch": 0.34493874919406836,
      "grad_norm": 7.9928202629089355,
      "kl": 0.106689453125,
      "learning_rate": 8.275306254029658e-07,
      "loss": 0.0043,
      "reward": 1.5217041969299316,
      "reward_std": 0.06747284345328808,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5217041969299316,
      "step": 535
    },
    {
      "completion_length": 115.75,
      "epoch": 0.34558349451966475,
      "grad_norm": 14.964887619018555,
      "kl": 0.109375,
      "learning_rate": 8.272082527401676e-07,
      "loss": 0.0044,
      "reward": 1.5932498574256897,
      "reward_std": 0.11327129602432251,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5932497978210449,
      "step": 536
    },
    {
      "completion_length": 108.9375,
      "epoch": 0.34622823984526113,
      "grad_norm": 13.483436584472656,
      "kl": 0.1064453125,
      "learning_rate": 8.268858800773694e-07,
      "loss": 0.0043,
      "reward": 1.4369852542877197,
      "reward_std": 0.10051735490560532,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4369853138923645,
      "step": 537
    },
    {
      "completion_length": 125.0625,
      "epoch": 0.3468729851708575,
      "grad_norm": 27.211544036865234,
      "kl": 0.11083984375,
      "learning_rate": 8.265635074145713e-07,
      "loss": 0.0044,
      "reward": 1.6145740747451782,
      "reward_std": 0.11309744045138359,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6145740449428558,
      "step": 538
    },
    {
      "completion_length": 121.9375,
      "epoch": 0.3475177304964539,
      "grad_norm": 11.06650447845459,
      "kl": 0.129150390625,
      "learning_rate": 8.26241134751773e-07,
      "loss": 0.0052,
      "reward": 1.518451452255249,
      "reward_std": 0.10257099196314812,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5184515118598938,
      "step": 539
    },
    {
      "completion_length": 120.546875,
      "epoch": 0.3481624758220503,
      "grad_norm": 27.978939056396484,
      "kl": 0.10546875,
      "learning_rate": 8.259187620889749e-07,
      "loss": 0.0042,
      "reward": 1.5805214643478394,
      "reward_std": 0.10257112607359886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5805214643478394,
      "step": 540
    },
    {
      "completion_length": 122.21875,
      "epoch": 0.3488072211476467,
      "grad_norm": 10.80872631072998,
      "kl": 0.109619140625,
      "learning_rate": 8.255963894261766e-07,
      "loss": 0.0044,
      "reward": 1.4624775052070618,
      "reward_std": 0.09696153551340103,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4624774754047394,
      "step": 541
    },
    {
      "completion_length": 109.28125,
      "epoch": 0.34945196647324306,
      "grad_norm": 12.653077125549316,
      "kl": 0.133544921875,
      "learning_rate": 8.252740167633784e-07,
      "loss": 0.0053,
      "reward": 1.588198721408844,
      "reward_std": 0.11155318468809128,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.588198721408844,
      "step": 542
    },
    {
      "completion_length": 120.09375,
      "epoch": 0.35009671179883944,
      "grad_norm": 22.150753021240234,
      "kl": 0.118408203125,
      "learning_rate": 8.249516441005803e-07,
      "loss": 0.0047,
      "reward": 1.56528639793396,
      "reward_std": 0.10561125725507736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.565286323428154,
      "step": 543
    },
    {
      "completion_length": 125.9375,
      "epoch": 0.35074145712443583,
      "grad_norm": 16.026426315307617,
      "kl": 0.119140625,
      "learning_rate": 8.246292714377821e-07,
      "loss": 0.0048,
      "reward": 1.427088737487793,
      "reward_std": 0.17949910461902618,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.44271376729011536,
      "step": 544
    },
    {
      "completion_length": 125.15625,
      "epoch": 0.3513862024500322,
      "grad_norm": 14.44304370880127,
      "kl": 0.11181640625,
      "learning_rate": 8.243068987749838e-07,
      "loss": 0.0045,
      "reward": 1.582537829875946,
      "reward_std": 0.09015891328454018,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5825379192829132,
      "step": 545
    },
    {
      "completion_length": 126.25,
      "epoch": 0.3520309477756286,
      "grad_norm": 17.329204559326172,
      "kl": 0.10498046875,
      "learning_rate": 8.239845261121857e-07,
      "loss": 0.0042,
      "reward": 1.6583821773529053,
      "reward_std": 0.13926228135824203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6583821475505829,
      "step": 546
    },
    {
      "completion_length": 123.859375,
      "epoch": 0.35267569310122504,
      "grad_norm": 24.32595443725586,
      "kl": 0.112548828125,
      "learning_rate": 8.236621534493875e-07,
      "loss": 0.0045,
      "reward": 1.613775610923767,
      "reward_std": 0.11163278669118881,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6137756109237671,
      "step": 547
    },
    {
      "completion_length": 120.5625,
      "epoch": 0.3533204384268214,
      "grad_norm": 13.765203475952148,
      "kl": 0.112548828125,
      "learning_rate": 8.233397807865893e-07,
      "loss": 0.0045,
      "reward": 1.637283205986023,
      "reward_std": 0.12223464995622635,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.637283205986023,
      "step": 548
    },
    {
      "completion_length": 128.75,
      "epoch": 0.3539651837524178,
      "grad_norm": 7.839471340179443,
      "kl": 0.140625,
      "learning_rate": 8.23017408123791e-07,
      "loss": 0.0056,
      "reward": 1.7445607781410217,
      "reward_std": 0.1211717426776886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7445608079433441,
      "step": 549
    },
    {
      "completion_length": 128.96875,
      "epoch": 0.3546099290780142,
      "grad_norm": 10.117485046386719,
      "kl": 0.11376953125,
      "learning_rate": 8.226950354609929e-07,
      "loss": 0.0045,
      "reward": 1.4675299525260925,
      "reward_std": 0.12889251857995987,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4675299674272537,
      "step": 550
    },
    {
      "completion_length": 114.296875,
      "epoch": 0.3552546744036106,
      "grad_norm": 26.474218368530273,
      "kl": 0.132568359375,
      "learning_rate": 8.223726627981947e-07,
      "loss": 0.0053,
      "reward": 1.5867825746536255,
      "reward_std": 0.10010011121630669,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5867825746536255,
      "step": 551
    },
    {
      "completion_length": 130.15625,
      "epoch": 0.35589941972920697,
      "grad_norm": 8.115386962890625,
      "kl": 0.109375,
      "learning_rate": 8.220502901353965e-07,
      "loss": 0.0044,
      "reward": 1.4267299175262451,
      "reward_std": 0.06233101151883602,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4267299324274063,
      "step": 552
    },
    {
      "completion_length": 130.5625,
      "epoch": 0.35654416505480335,
      "grad_norm": 31.162986755371094,
      "kl": 0.125244140625,
      "learning_rate": 8.217279174725983e-07,
      "loss": 0.005,
      "reward": 1.4682827591896057,
      "reward_std": 0.19038020819425583,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.48390769958496094,
      "step": 553
    },
    {
      "completion_length": 123.96875,
      "epoch": 0.35718891038039974,
      "grad_norm": 9.136853218078613,
      "kl": 0.113525390625,
      "learning_rate": 8.214055448098001e-07,
      "loss": 0.0045,
      "reward": 1.6539664268493652,
      "reward_std": 0.09818685799837112,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6539664268493652,
      "step": 554
    },
    {
      "completion_length": 122.3125,
      "epoch": 0.3578336557059961,
      "grad_norm": 13.743880271911621,
      "kl": 0.11962890625,
      "learning_rate": 8.210831721470018e-07,
      "loss": 0.0048,
      "reward": 1.4207700490951538,
      "reward_std": 0.11169535666704178,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4207700490951538,
      "step": 555
    },
    {
      "completion_length": 132.734375,
      "epoch": 0.3584784010315925,
      "grad_norm": 19.605443954467773,
      "kl": 0.12841796875,
      "learning_rate": 8.207607994842038e-07,
      "loss": 0.0051,
      "reward": 1.6942888498306274,
      "reward_std": 0.1028636246919632,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6942889094352722,
      "step": 556
    },
    {
      "completion_length": 130.140625,
      "epoch": 0.3591231463571889,
      "grad_norm": 10.357660293579102,
      "kl": 0.107666015625,
      "learning_rate": 8.204384268214055e-07,
      "loss": 0.0043,
      "reward": 1.49127995967865,
      "reward_std": 0.11323336511850357,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49128003418445587,
      "step": 557
    },
    {
      "completion_length": 149.734375,
      "epoch": 0.3597678916827853,
      "grad_norm": 8.94410514831543,
      "kl": 0.1181640625,
      "learning_rate": 8.201160541586073e-07,
      "loss": 0.0047,
      "reward": 1.739545226097107,
      "reward_std": 0.0977691225707531,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7395451962947845,
      "step": 558
    },
    {
      "completion_length": 142.34375,
      "epoch": 0.36041263700838166,
      "grad_norm": 9.473371505737305,
      "kl": 0.109619140625,
      "learning_rate": 8.197936814958091e-07,
      "loss": 0.0044,
      "reward": 1.619294822216034,
      "reward_std": 0.10096556693315506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6192947924137115,
      "step": 559
    },
    {
      "completion_length": 124.953125,
      "epoch": 0.3610573823339781,
      "grad_norm": 14.380293846130371,
      "kl": 0.1201171875,
      "learning_rate": 8.19471308833011e-07,
      "loss": 0.0048,
      "reward": 1.418857455253601,
      "reward_std": 0.08556991256773472,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41885746270418167,
      "step": 560
    },
    {
      "completion_length": 136.6875,
      "epoch": 0.3617021276595745,
      "grad_norm": 9.324554443359375,
      "kl": 0.114013671875,
      "learning_rate": 8.191489361702127e-07,
      "loss": 0.0046,
      "reward": 1.5292004346847534,
      "reward_std": 0.07976378127932549,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5292005091905594,
      "step": 561
    },
    {
      "completion_length": 155.71875,
      "epoch": 0.3623468729851709,
      "grad_norm": 10.505910873413086,
      "kl": 0.131103515625,
      "learning_rate": 8.188265635074146e-07,
      "loss": 0.0053,
      "reward": 1.5651956796646118,
      "reward_std": 0.11247077211737633,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5651956498622894,
      "step": 562
    },
    {
      "completion_length": 146.3125,
      "epoch": 0.36299161831076726,
      "grad_norm": 9.546722412109375,
      "kl": 0.1083984375,
      "learning_rate": 8.185041908446163e-07,
      "loss": 0.0043,
      "reward": 1.4872339367866516,
      "reward_std": 0.09367235377430916,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4872339963912964,
      "step": 563
    },
    {
      "completion_length": 131.546875,
      "epoch": 0.36363636363636365,
      "grad_norm": 13.24105453491211,
      "kl": 0.130859375,
      "learning_rate": 8.181818181818182e-07,
      "loss": 0.0052,
      "reward": 1.458554744720459,
      "reward_std": 0.10819820687174797,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.458554744720459,
      "step": 564
    },
    {
      "completion_length": 149.265625,
      "epoch": 0.36428110896196003,
      "grad_norm": 7.181205749511719,
      "kl": 0.122802734375,
      "learning_rate": 8.1785944551902e-07,
      "loss": 0.0049,
      "reward": 1.4140058159828186,
      "reward_std": 0.10015027970075607,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4140057861804962,
      "step": 565
    },
    {
      "completion_length": 139.8125,
      "epoch": 0.3649258542875564,
      "grad_norm": 13.35418701171875,
      "kl": 0.197265625,
      "learning_rate": 8.175370728562218e-07,
      "loss": 0.0079,
      "reward": 1.561979353427887,
      "reward_std": 0.21178478002548218,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.577604353427887,
      "step": 566
    },
    {
      "completion_length": 128.84375,
      "epoch": 0.3655705996131528,
      "grad_norm": 51.9110107421875,
      "kl": 0.114013671875,
      "learning_rate": 8.172147001934235e-07,
      "loss": 0.0046,
      "reward": 1.7140253782272339,
      "reward_std": 0.16923753917217255,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7140253484249115,
      "step": 567
    },
    {
      "completion_length": 135.4375,
      "epoch": 0.3662153449387492,
      "grad_norm": 8.429132461547852,
      "kl": 0.1259765625,
      "learning_rate": 8.168923275306254e-07,
      "loss": 0.005,
      "reward": 1.4810205698013306,
      "reward_std": 0.10634732991456985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4810205101966858,
      "step": 568
    },
    {
      "completion_length": 142.625,
      "epoch": 0.3668600902643456,
      "grad_norm": 11.647993087768555,
      "kl": 0.12451171875,
      "learning_rate": 8.165699548678272e-07,
      "loss": 0.005,
      "reward": 1.3584343194961548,
      "reward_std": 0.07176856696605682,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3584343045949936,
      "step": 569
    },
    {
      "completion_length": 129.375,
      "epoch": 0.36750483558994196,
      "grad_norm": 7.475222110748291,
      "kl": 0.13818359375,
      "learning_rate": 8.16247582205029e-07,
      "loss": 0.0055,
      "reward": 1.5198262929916382,
      "reward_std": 0.12593180686235428,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.519826278090477,
      "step": 570
    },
    {
      "completion_length": 121.375,
      "epoch": 0.36814958091553834,
      "grad_norm": 10.700190544128418,
      "kl": 0.123046875,
      "learning_rate": 8.159252095422307e-07,
      "loss": 0.0049,
      "reward": 1.5901803374290466,
      "reward_std": 0.1543055735528469,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.590180367231369,
      "step": 571
    },
    {
      "completion_length": 129.015625,
      "epoch": 0.36879432624113473,
      "grad_norm": 11.171639442443848,
      "kl": 0.111083984375,
      "learning_rate": 8.156028368794326e-07,
      "loss": 0.0044,
      "reward": 1.3946861624717712,
      "reward_std": 0.11192984879016876,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.39468614757061005,
      "step": 572
    },
    {
      "completion_length": 124.875,
      "epoch": 0.3694390715667311,
      "grad_norm": 8.263480186462402,
      "kl": 0.1396484375,
      "learning_rate": 8.152804642166344e-07,
      "loss": 0.0056,
      "reward": 1.6307023167610168,
      "reward_std": 0.18546786159276962,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.6775772869586945,
      "step": 573
    },
    {
      "completion_length": 123.46875,
      "epoch": 0.37008381689232756,
      "grad_norm": 26.08427619934082,
      "kl": 0.1220703125,
      "learning_rate": 8.149580915538362e-07,
      "loss": 0.0049,
      "reward": 1.4789862036705017,
      "reward_std": 0.15343213826417923,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4946112334728241,
      "step": 574
    },
    {
      "completion_length": 135.953125,
      "epoch": 0.37072856221792394,
      "grad_norm": 10.028849601745605,
      "kl": 0.114990234375,
      "learning_rate": 8.14635718891038e-07,
      "loss": 0.0046,
      "reward": 1.5295458436012268,
      "reward_std": 0.12414504960179329,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5295458734035492,
      "step": 575
    },
    {
      "completion_length": 128.015625,
      "epoch": 0.3713733075435203,
      "grad_norm": 13.812504768371582,
      "kl": 0.112060546875,
      "learning_rate": 8.143133462282398e-07,
      "loss": 0.0045,
      "reward": 1.5613334774971008,
      "reward_std": 0.10750623792409897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5613334625959396,
      "step": 576
    },
    {
      "completion_length": 149.625,
      "epoch": 0.3720180528691167,
      "grad_norm": 27.1822509765625,
      "kl": 0.102294921875,
      "learning_rate": 8.139909735654416e-07,
      "loss": 0.0041,
      "reward": 1.6229792833328247,
      "reward_std": 0.11508912220597267,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6229793429374695,
      "step": 577
    },
    {
      "completion_length": 124.609375,
      "epoch": 0.3726627981947131,
      "grad_norm": 9.831287384033203,
      "kl": 0.12109375,
      "learning_rate": 8.136686009026435e-07,
      "loss": 0.0048,
      "reward": 1.601158857345581,
      "reward_std": 0.11256638541817665,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.601158857345581,
      "step": 578
    },
    {
      "completion_length": 134.984375,
      "epoch": 0.3733075435203095,
      "grad_norm": 26.67717170715332,
      "kl": 0.1416015625,
      "learning_rate": 8.133462282398452e-07,
      "loss": 0.0057,
      "reward": 1.4907135963439941,
      "reward_std": 0.1917910873889923,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5063386261463165,
      "step": 579
    },
    {
      "completion_length": 131.421875,
      "epoch": 0.37395228884590587,
      "grad_norm": 6.938395023345947,
      "kl": 0.111328125,
      "learning_rate": 8.13023855577047e-07,
      "loss": 0.0045,
      "reward": 1.559043526649475,
      "reward_std": 0.10559235885739326,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5590435266494751,
      "step": 580
    },
    {
      "completion_length": 138.9375,
      "epoch": 0.37459703417150225,
      "grad_norm": 21.588024139404297,
      "kl": 0.113525390625,
      "learning_rate": 8.127014829142488e-07,
      "loss": 0.0045,
      "reward": 1.5905563831329346,
      "reward_std": 0.09706471487879753,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5905563831329346,
      "step": 581
    },
    {
      "completion_length": 134.515625,
      "epoch": 0.37524177949709864,
      "grad_norm": 17.881168365478516,
      "kl": 0.119140625,
      "learning_rate": 8.123791102514507e-07,
      "loss": 0.0048,
      "reward": 1.543268859386444,
      "reward_std": 0.18078606575727463,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5432688891887665,
      "step": 582
    },
    {
      "completion_length": 145.953125,
      "epoch": 0.375886524822695,
      "grad_norm": 29.037656784057617,
      "kl": 0.10107421875,
      "learning_rate": 8.120567375886524e-07,
      "loss": 0.004,
      "reward": 1.6184459328651428,
      "reward_std": 0.05552523583173752,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6184459626674652,
      "step": 583
    },
    {
      "completion_length": 145.234375,
      "epoch": 0.3765312701482914,
      "grad_norm": 15.09101676940918,
      "kl": 0.13525390625,
      "learning_rate": 8.117343649258543e-07,
      "loss": 0.0054,
      "reward": 1.487252116203308,
      "reward_std": 0.1185568068176508,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4872521460056305,
      "step": 584
    },
    {
      "completion_length": 125.8125,
      "epoch": 0.3771760154738878,
      "grad_norm": 11.656231880187988,
      "kl": 0.1201171875,
      "learning_rate": 8.11411992263056e-07,
      "loss": 0.0048,
      "reward": 1.4543855786323547,
      "reward_std": 0.09149051457643509,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45438555628061295,
      "step": 585
    },
    {
      "completion_length": 134.234375,
      "epoch": 0.3778207607994842,
      "grad_norm": 28.694446563720703,
      "kl": 0.100341796875,
      "learning_rate": 8.110896196002579e-07,
      "loss": 0.004,
      "reward": 1.495825171470642,
      "reward_std": 0.14905067533254623,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4958251863718033,
      "step": 586
    },
    {
      "completion_length": 125.84375,
      "epoch": 0.3784655061250806,
      "grad_norm": 12.141417503356934,
      "kl": 0.136962890625,
      "learning_rate": 8.107672469374597e-07,
      "loss": 0.0055,
      "reward": 1.6799025535583496,
      "reward_std": 0.15780113637447357,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6799026131629944,
      "step": 587
    },
    {
      "completion_length": 145.375,
      "epoch": 0.379110251450677,
      "grad_norm": 8.39338207244873,
      "kl": 0.1220703125,
      "learning_rate": 8.104448742746615e-07,
      "loss": 0.0049,
      "reward": 1.6171015501022339,
      "reward_std": 0.20077940821647644,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6327264904975891,
      "step": 588
    },
    {
      "completion_length": 125.796875,
      "epoch": 0.3797549967762734,
      "grad_norm": 10.23801040649414,
      "kl": 0.110595703125,
      "learning_rate": 8.101225016118632e-07,
      "loss": 0.0044,
      "reward": 1.3968492150306702,
      "reward_std": 0.11241466552019119,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3968491703271866,
      "step": 589
    },
    {
      "completion_length": 137.03125,
      "epoch": 0.3803997421018698,
      "grad_norm": 11.222281455993652,
      "kl": 0.142578125,
      "learning_rate": 8.098001289490652e-07,
      "loss": 0.0057,
      "reward": 1.656241774559021,
      "reward_std": 0.11266876012086868,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6562418043613434,
      "step": 590
    },
    {
      "completion_length": 133.859375,
      "epoch": 0.38104448742746616,
      "grad_norm": 14.622573852539062,
      "kl": 0.112060546875,
      "learning_rate": 8.094777562862669e-07,
      "loss": 0.0045,
      "reward": 1.6198719143867493,
      "reward_std": 0.14295167103409767,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6198718845844269,
      "step": 591
    },
    {
      "completion_length": 138.921875,
      "epoch": 0.38168923275306255,
      "grad_norm": 14.126190185546875,
      "kl": 0.124267578125,
      "learning_rate": 8.091553836234687e-07,
      "loss": 0.005,
      "reward": 1.544808268547058,
      "reward_std": 0.16154202073812485,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5448082685470581,
      "step": 592
    },
    {
      "completion_length": 140.140625,
      "epoch": 0.38233397807865893,
      "grad_norm": 10.186321258544922,
      "kl": 0.10791015625,
      "learning_rate": 8.088330109606704e-07,
      "loss": 0.0043,
      "reward": 1.633695662021637,
      "reward_std": 0.09759926423430443,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.633695662021637,
      "step": 593
    },
    {
      "completion_length": 129.8125,
      "epoch": 0.3829787234042553,
      "grad_norm": 13.927589416503906,
      "kl": 0.108642578125,
      "learning_rate": 8.085106382978723e-07,
      "loss": 0.0043,
      "reward": 1.5086371898651123,
      "reward_std": 0.12473899126052856,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5086371302604675,
      "step": 594
    },
    {
      "completion_length": 125.625,
      "epoch": 0.3836234687298517,
      "grad_norm": 13.513331413269043,
      "kl": 0.119140625,
      "learning_rate": 8.081882656350741e-07,
      "loss": 0.0048,
      "reward": 1.7253805994987488,
      "reward_std": 0.16587653756141663,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7410056293010712,
      "step": 595
    },
    {
      "completion_length": 135.125,
      "epoch": 0.3842682140554481,
      "grad_norm": 9.143721580505371,
      "kl": 0.10986328125,
      "learning_rate": 8.07865892972276e-07,
      "loss": 0.0044,
      "reward": 1.5537520051002502,
      "reward_std": 0.09487171098589897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5537519752979279,
      "step": 596
    },
    {
      "completion_length": 126.421875,
      "epoch": 0.3849129593810445,
      "grad_norm": 10.566431045532227,
      "kl": 0.10302734375,
      "learning_rate": 8.075435203094777e-07,
      "loss": 0.0041,
      "reward": 1.5097746849060059,
      "reward_std": 0.07031143084168434,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5097746551036835,
      "step": 597
    },
    {
      "completion_length": 130.84375,
      "epoch": 0.38555770470664086,
      "grad_norm": 8.138565063476562,
      "kl": 0.119140625,
      "learning_rate": 8.072211476466795e-07,
      "loss": 0.0048,
      "reward": 1.6204659342765808,
      "reward_std": 0.1338505893945694,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6360909044742584,
      "step": 598
    },
    {
      "completion_length": 136.28125,
      "epoch": 0.38620245003223724,
      "grad_norm": 10.86203670501709,
      "kl": 0.11865234375,
      "learning_rate": 8.068987749838813e-07,
      "loss": 0.0047,
      "reward": 1.634830355644226,
      "reward_std": 0.08793558552861214,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6348303556442261,
      "step": 599
    },
    {
      "completion_length": 130.4375,
      "epoch": 0.38684719535783363,
      "grad_norm": 7.916982650756836,
      "kl": 0.11376953125,
      "learning_rate": 8.065764023210832e-07,
      "loss": 0.0045,
      "reward": 1.5632728338241577,
      "reward_std": 0.10517224669456482,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5632728040218353,
      "step": 600
    },
    {
      "completion_length": 155.3125,
      "epoch": 0.38749194068343007,
      "grad_norm": 13.323336601257324,
      "kl": 0.11083984375,
      "learning_rate": 8.062540296582849e-07,
      "loss": 0.0044,
      "reward": 1.4888296127319336,
      "reward_std": 0.11279592663049698,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4888295754790306,
      "step": 601
    },
    {
      "completion_length": 134.046875,
      "epoch": 0.38813668600902645,
      "grad_norm": 11.745083808898926,
      "kl": 0.11767578125,
      "learning_rate": 8.059316569954867e-07,
      "loss": 0.0047,
      "reward": 1.597053587436676,
      "reward_std": 0.1300004981458187,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.597053587436676,
      "step": 602
    },
    {
      "completion_length": 145.640625,
      "epoch": 0.38878143133462284,
      "grad_norm": 8.197556495666504,
      "kl": 0.1171875,
      "learning_rate": 8.056092843326886e-07,
      "loss": 0.0047,
      "reward": 1.5338327884674072,
      "reward_std": 0.15634901821613312,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5338327586650848,
      "step": 603
    },
    {
      "completion_length": 142.859375,
      "epoch": 0.3894261766602192,
      "grad_norm": 19.091354370117188,
      "kl": 0.111572265625,
      "learning_rate": 8.052869116698904e-07,
      "loss": 0.0045,
      "reward": 1.6582316160202026,
      "reward_std": 0.11573278903961182,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6582316756248474,
      "step": 604
    },
    {
      "completion_length": 147.296875,
      "epoch": 0.3900709219858156,
      "grad_norm": 28.295654296875,
      "kl": 0.128662109375,
      "learning_rate": 8.049645390070921e-07,
      "loss": 0.0051,
      "reward": 1.4338959455490112,
      "reward_std": 0.10549728572368622,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43389593064785004,
      "step": 605
    },
    {
      "completion_length": 137.96875,
      "epoch": 0.390715667311412,
      "grad_norm": 137.10951232910156,
      "kl": 0.112060546875,
      "learning_rate": 8.04642166344294e-07,
      "loss": 0.0045,
      "reward": 1.6279088258743286,
      "reward_std": 0.1394149400293827,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6279088854789734,
      "step": 606
    },
    {
      "completion_length": 132.046875,
      "epoch": 0.3913604126370084,
      "grad_norm": 18.1970272064209,
      "kl": 0.13427734375,
      "learning_rate": 8.043197936814957e-07,
      "loss": 0.0054,
      "reward": 1.4707345366477966,
      "reward_std": 0.10406222194433212,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47073453664779663,
      "step": 607
    },
    {
      "completion_length": 147.9375,
      "epoch": 0.39200515796260477,
      "grad_norm": 9.023652076721191,
      "kl": 0.112060546875,
      "learning_rate": 8.039974210186976e-07,
      "loss": 0.0045,
      "reward": 1.6449916362762451,
      "reward_std": 0.22956346720457077,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6606166958808899,
      "step": 608
    },
    {
      "completion_length": 134.5,
      "epoch": 0.39264990328820115,
      "grad_norm": 12.73431396484375,
      "kl": 0.102783203125,
      "learning_rate": 8.036750483558994e-07,
      "loss": 0.0041,
      "reward": 1.4050042033195496,
      "reward_std": 0.09452563524246216,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4050041437149048,
      "step": 609
    },
    {
      "completion_length": 123.96875,
      "epoch": 0.39329464861379754,
      "grad_norm": 15.1465425491333,
      "kl": 0.11572265625,
      "learning_rate": 8.033526756931012e-07,
      "loss": 0.0046,
      "reward": 1.5025982856750488,
      "reward_std": 0.1606716513633728,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5182233303785324,
      "step": 610
    },
    {
      "completion_length": 128.84375,
      "epoch": 0.3939393939393939,
      "grad_norm": 10.938918113708496,
      "kl": 0.11865234375,
      "learning_rate": 8.030303030303029e-07,
      "loss": 0.0047,
      "reward": 1.56488835811615,
      "reward_std": 0.10107577964663506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5648883879184723,
      "step": 611
    },
    {
      "completion_length": 130.546875,
      "epoch": 0.3945841392649903,
      "grad_norm": 10.713906288146973,
      "kl": 0.2900390625,
      "learning_rate": 8.027079303675049e-07,
      "loss": 0.0116,
      "reward": 1.6501939296722412,
      "reward_std": 0.1356111727654934,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6501939296722412,
      "step": 612
    },
    {
      "completion_length": 129.109375,
      "epoch": 0.3952288845905867,
      "grad_norm": 12.522553443908691,
      "kl": 0.101806640625,
      "learning_rate": 8.023855577047066e-07,
      "loss": 0.0041,
      "reward": 1.6621387004852295,
      "reward_std": 0.09056547284126282,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6621387004852295,
      "step": 613
    },
    {
      "completion_length": 114.875,
      "epoch": 0.39587362991618313,
      "grad_norm": 14.493385314941406,
      "kl": 0.108642578125,
      "learning_rate": 8.020631850419084e-07,
      "loss": 0.0043,
      "reward": 1.5702993869781494,
      "reward_std": 0.09657874330878258,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5702993869781494,
      "step": 614
    },
    {
      "completion_length": 130.0625,
      "epoch": 0.3965183752417795,
      "grad_norm": 14.518086433410645,
      "kl": 0.092041015625,
      "learning_rate": 8.017408123791101e-07,
      "loss": 0.0037,
      "reward": 1.5168561935424805,
      "reward_std": 0.12171942368149757,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5168561637401581,
      "step": 615
    },
    {
      "completion_length": 114.875,
      "epoch": 0.3971631205673759,
      "grad_norm": 13.475616455078125,
      "kl": 0.10693359375,
      "learning_rate": 8.014184397163121e-07,
      "loss": 0.0043,
      "reward": 1.5117144584655762,
      "reward_std": 0.16558092087507248,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5117143988609314,
      "step": 616
    },
    {
      "completion_length": 132.4375,
      "epoch": 0.3978078658929723,
      "grad_norm": 11.467823028564453,
      "kl": 0.097900390625,
      "learning_rate": 8.010960670535138e-07,
      "loss": 0.0039,
      "reward": 1.4913342595100403,
      "reward_std": 0.17832869291305542,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5069593191146851,
      "step": 617
    },
    {
      "completion_length": 115.921875,
      "epoch": 0.3984526112185687,
      "grad_norm": 14.274715423583984,
      "kl": 0.107666015625,
      "learning_rate": 8.007736943907156e-07,
      "loss": 0.0043,
      "reward": 1.3526421785354614,
      "reward_std": 0.07542551681399345,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3526422083377838,
      "step": 618
    },
    {
      "completion_length": 116.875,
      "epoch": 0.39909735654416506,
      "grad_norm": 9.649855613708496,
      "kl": 0.1298828125,
      "learning_rate": 8.004513217279174e-07,
      "loss": 0.0052,
      "reward": 1.565634310245514,
      "reward_std": 0.1185467280447483,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5656343698501587,
      "step": 619
    },
    {
      "completion_length": 118.9375,
      "epoch": 0.39974210186976145,
      "grad_norm": 22.857391357421875,
      "kl": 0.133544921875,
      "learning_rate": 8.001289490651193e-07,
      "loss": 0.0053,
      "reward": 1.5444243550300598,
      "reward_std": 0.09491568803787231,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5444244742393494,
      "step": 620
    },
    {
      "completion_length": 119.453125,
      "epoch": 0.40038684719535783,
      "grad_norm": 20.537433624267578,
      "kl": 0.114013671875,
      "learning_rate": 7.99806576402321e-07,
      "loss": 0.0046,
      "reward": 1.4165579080581665,
      "reward_std": 0.16316518932580948,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4321829378604889,
      "step": 621
    },
    {
      "completion_length": 115.0625,
      "epoch": 0.4010315925209542,
      "grad_norm": 22.098909378051758,
      "kl": 0.1640625,
      "learning_rate": 7.994842037395229e-07,
      "loss": 0.0066,
      "reward": 1.5407551527023315,
      "reward_std": 0.17422685772180557,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5407552272081375,
      "step": 622
    },
    {
      "completion_length": 118.734375,
      "epoch": 0.4016763378465506,
      "grad_norm": 48.7105827331543,
      "kl": 0.107421875,
      "learning_rate": 7.991618310767246e-07,
      "loss": 0.0043,
      "reward": 1.6025624871253967,
      "reward_std": 0.10630748048424721,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.602562427520752,
      "step": 623
    },
    {
      "completion_length": 119.59375,
      "epoch": 0.402321083172147,
      "grad_norm": 17.388961791992188,
      "kl": 0.095458984375,
      "learning_rate": 7.988394584139264e-07,
      "loss": 0.0038,
      "reward": 1.6318548321723938,
      "reward_std": 0.1776801347732544,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6318548321723938,
      "step": 624
    },
    {
      "completion_length": 121.046875,
      "epoch": 0.4029658284977434,
      "grad_norm": 7.685291290283203,
      "kl": 0.110107421875,
      "learning_rate": 7.985170857511283e-07,
      "loss": 0.0044,
      "reward": 1.5066440105438232,
      "reward_std": 0.10979078710079193,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.506644070148468,
      "step": 625
    },
    {
      "completion_length": 123.765625,
      "epoch": 0.40361057382333976,
      "grad_norm": 20.131128311157227,
      "kl": 0.11767578125,
      "learning_rate": 7.981947130883301e-07,
      "loss": 0.0047,
      "reward": 1.6256684064865112,
      "reward_std": 0.1347283013164997,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6256683468818665,
      "step": 626
    },
    {
      "completion_length": 112.890625,
      "epoch": 0.40425531914893614,
      "grad_norm": 7.655572891235352,
      "kl": 0.119873046875,
      "learning_rate": 7.978723404255318e-07,
      "loss": 0.0048,
      "reward": 1.525473415851593,
      "reward_std": 0.1499604694545269,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5254734456539154,
      "step": 627
    },
    {
      "completion_length": 130.328125,
      "epoch": 0.4049000644745326,
      "grad_norm": 95.7166519165039,
      "kl": 0.107666015625,
      "learning_rate": 7.975499677627337e-07,
      "loss": 0.0043,
      "reward": 1.5183069109916687,
      "reward_std": 0.12149505317211151,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5183069556951523,
      "step": 628
    },
    {
      "completion_length": 136.890625,
      "epoch": 0.40554480980012897,
      "grad_norm": 16.36639404296875,
      "kl": 0.10693359375,
      "learning_rate": 7.972275950999355e-07,
      "loss": 0.0043,
      "reward": 1.6043989658355713,
      "reward_std": 0.1741345077753067,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6043989807367325,
      "step": 629
    },
    {
      "completion_length": 132.0625,
      "epoch": 0.40618955512572535,
      "grad_norm": 9.136812210083008,
      "kl": 0.10986328125,
      "learning_rate": 7.969052224371373e-07,
      "loss": 0.0044,
      "reward": 1.3735297918319702,
      "reward_std": 0.13789300620555878,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.373529776930809,
      "step": 630
    },
    {
      "completion_length": 115.65625,
      "epoch": 0.40683430045132174,
      "grad_norm": 7.693540573120117,
      "kl": 0.1201171875,
      "learning_rate": 7.965828497743391e-07,
      "loss": 0.0048,
      "reward": 1.6083815693855286,
      "reward_std": 0.1517021358013153,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6083815693855286,
      "step": 631
    },
    {
      "completion_length": 125.515625,
      "epoch": 0.4074790457769181,
      "grad_norm": 7.784499645233154,
      "kl": 0.11962890625,
      "learning_rate": 7.962604771115409e-07,
      "loss": 0.0048,
      "reward": 1.487956166267395,
      "reward_std": 0.16370031237602234,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48795612156391144,
      "step": 632
    },
    {
      "completion_length": 108.140625,
      "epoch": 0.4081237911025145,
      "grad_norm": 6.592469215393066,
      "kl": 0.123779296875,
      "learning_rate": 7.959381044487427e-07,
      "loss": 0.005,
      "reward": 1.7070671916007996,
      "reward_std": 0.08668256551027298,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7070672810077667,
      "step": 633
    },
    {
      "completion_length": 113.203125,
      "epoch": 0.4087685364281109,
      "grad_norm": 10.177643775939941,
      "kl": 0.113525390625,
      "learning_rate": 7.956157317859446e-07,
      "loss": 0.0045,
      "reward": 1.6079189777374268,
      "reward_std": 0.12942004203796387,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6079188883304596,
      "step": 634
    },
    {
      "completion_length": 120.109375,
      "epoch": 0.4094132817537073,
      "grad_norm": 8.311883926391602,
      "kl": 0.1103515625,
      "learning_rate": 7.952933591231463e-07,
      "loss": 0.0044,
      "reward": 1.634382724761963,
      "reward_std": 0.19498103857040405,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.63438281416893,
      "step": 635
    },
    {
      "completion_length": 105.84375,
      "epoch": 0.41005802707930367,
      "grad_norm": 15.116155624389648,
      "kl": 0.104736328125,
      "learning_rate": 7.949709864603481e-07,
      "loss": 0.0042,
      "reward": 1.5293989777565002,
      "reward_std": 0.11750294640660286,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5293989479541779,
      "step": 636
    },
    {
      "completion_length": 103.53125,
      "epoch": 0.41070277240490005,
      "grad_norm": 8.729680061340332,
      "kl": 0.11474609375,
      "learning_rate": 7.946486137975498e-07,
      "loss": 0.0046,
      "reward": 1.5659652352333069,
      "reward_std": 0.15115171670913696,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5659652352333069,
      "step": 637
    },
    {
      "completion_length": 108.03125,
      "epoch": 0.41134751773049644,
      "grad_norm": 47.11723327636719,
      "kl": 0.1240234375,
      "learning_rate": 7.943262411347518e-07,
      "loss": 0.005,
      "reward": 1.554872214794159,
      "reward_std": 0.1266670562326908,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5548722743988037,
      "step": 638
    },
    {
      "completion_length": 114.296875,
      "epoch": 0.4119922630560928,
      "grad_norm": 9.169736862182617,
      "kl": 0.1259765625,
      "learning_rate": 7.940038684719535e-07,
      "loss": 0.005,
      "reward": 1.306950032711029,
      "reward_std": 0.07547219470143318,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.30694999545812607,
      "step": 639
    },
    {
      "completion_length": 107.890625,
      "epoch": 0.4126370083816892,
      "grad_norm": 9.10857105255127,
      "kl": 0.10693359375,
      "learning_rate": 7.936814958091553e-07,
      "loss": 0.0043,
      "reward": 1.6737666726112366,
      "reward_std": 0.0729268491268158,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6737666428089142,
      "step": 640
    },
    {
      "completion_length": 107.0625,
      "epoch": 0.41328175370728565,
      "grad_norm": 21.960092544555664,
      "kl": 0.1416015625,
      "learning_rate": 7.933591231463571e-07,
      "loss": 0.0057,
      "reward": 1.5458187460899353,
      "reward_std": 0.13090715557336807,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5458187162876129,
      "step": 641
    },
    {
      "completion_length": 118.515625,
      "epoch": 0.41392649903288203,
      "grad_norm": 11.414142608642578,
      "kl": 0.140625,
      "learning_rate": 7.93036750483559e-07,
      "loss": 0.0056,
      "reward": 1.5467508435249329,
      "reward_std": 0.1591968908905983,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5623758435249329,
      "step": 642
    },
    {
      "completion_length": 103.515625,
      "epoch": 0.4145712443584784,
      "grad_norm": 8.753649711608887,
      "kl": 0.11572265625,
      "learning_rate": 7.927143778207607e-07,
      "loss": 0.0046,
      "reward": 1.463947594165802,
      "reward_std": 0.10150327160954475,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.463947594165802,
      "step": 643
    },
    {
      "completion_length": 110.328125,
      "epoch": 0.4152159896840748,
      "grad_norm": 16.589797973632812,
      "kl": 0.12060546875,
      "learning_rate": 7.923920051579626e-07,
      "loss": 0.0048,
      "reward": 1.4456722736358643,
      "reward_std": 0.16995223611593246,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.46129727363586426,
      "step": 644
    },
    {
      "completion_length": 109.5625,
      "epoch": 0.4158607350096712,
      "grad_norm": 11.583199501037598,
      "kl": 0.1083984375,
      "learning_rate": 7.920696324951643e-07,
      "loss": 0.0043,
      "reward": 1.4806134700775146,
      "reward_std": 0.1219950020313263,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48061345517635345,
      "step": 645
    },
    {
      "completion_length": 111.65625,
      "epoch": 0.4165054803352676,
      "grad_norm": 22.0245418548584,
      "kl": 0.11181640625,
      "learning_rate": 7.917472598323662e-07,
      "loss": 0.0045,
      "reward": 1.457335352897644,
      "reward_std": 0.07867556810379028,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45733533799648285,
      "step": 646
    },
    {
      "completion_length": 100.890625,
      "epoch": 0.41715022566086396,
      "grad_norm": 25.137718200683594,
      "kl": 0.14208984375,
      "learning_rate": 7.91424887169568e-07,
      "loss": 0.0057,
      "reward": 1.6399447917938232,
      "reward_std": 0.1750987023115158,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6555697917938232,
      "step": 647
    },
    {
      "completion_length": 115.546875,
      "epoch": 0.41779497098646035,
      "grad_norm": 23.508705139160156,
      "kl": 0.122314453125,
      "learning_rate": 7.911025145067698e-07,
      "loss": 0.0049,
      "reward": 1.4861845970153809,
      "reward_std": 0.052020519971847534,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4861845374107361,
      "step": 648
    },
    {
      "completion_length": 106.359375,
      "epoch": 0.41843971631205673,
      "grad_norm": 11.12207317352295,
      "kl": 0.11083984375,
      "learning_rate": 7.907801418439715e-07,
      "loss": 0.0044,
      "reward": 1.5252867937088013,
      "reward_std": 0.19501442462205887,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5409118235111237,
      "step": 649
    },
    {
      "completion_length": 104.046875,
      "epoch": 0.4190844616376531,
      "grad_norm": 10.382707595825195,
      "kl": 0.12890625,
      "learning_rate": 7.904577691811734e-07,
      "loss": 0.0052,
      "reward": 1.5812249779701233,
      "reward_std": 0.15144172310829163,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5812250077724457,
      "step": 650
    },
    {
      "completion_length": 104.640625,
      "epoch": 0.4197292069632495,
      "grad_norm": 27.51158332824707,
      "kl": 0.11572265625,
      "learning_rate": 7.901353965183752e-07,
      "loss": 0.0046,
      "reward": 1.6362160444259644,
      "reward_std": 0.14013683423399925,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6362160742282867,
      "step": 651
    },
    {
      "completion_length": 104.59375,
      "epoch": 0.4203739522888459,
      "grad_norm": 16.1353816986084,
      "kl": 0.1357421875,
      "learning_rate": 7.89813023855577e-07,
      "loss": 0.0054,
      "reward": 1.453727900981903,
      "reward_std": 0.12474618852138519,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4537279009819031,
      "step": 652
    },
    {
      "completion_length": 109.765625,
      "epoch": 0.42101869761444227,
      "grad_norm": 10.052741050720215,
      "kl": 0.131591796875,
      "learning_rate": 7.894906511927788e-07,
      "loss": 0.0053,
      "reward": 1.5141014456748962,
      "reward_std": 0.09088888391852379,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5141013860702515,
      "step": 653
    },
    {
      "completion_length": 105.71875,
      "epoch": 0.42166344294003866,
      "grad_norm": 13.285776138305664,
      "kl": 0.1083984375,
      "learning_rate": 7.891682785299806e-07,
      "loss": 0.0043,
      "reward": 1.5992873907089233,
      "reward_std": 0.1335230953991413,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5992873311042786,
      "step": 654
    },
    {
      "completion_length": 116.6875,
      "epoch": 0.4223081882656351,
      "grad_norm": 14.342023849487305,
      "kl": 0.138427734375,
      "learning_rate": 7.888459058671824e-07,
      "loss": 0.0055,
      "reward": 1.3910069465637207,
      "reward_std": 0.09431485831737518,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3910069465637207,
      "step": 655
    },
    {
      "completion_length": 98.203125,
      "epoch": 0.4229529335912315,
      "grad_norm": 13.409884452819824,
      "kl": 0.104248046875,
      "learning_rate": 7.885235332043843e-07,
      "loss": 0.0042,
      "reward": 1.4601246118545532,
      "reward_std": 0.07733478024601936,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46012458205223083,
      "step": 656
    },
    {
      "completion_length": 99.015625,
      "epoch": 0.42359767891682787,
      "grad_norm": 13.980112075805664,
      "kl": 0.135498046875,
      "learning_rate": 7.88201160541586e-07,
      "loss": 0.0054,
      "reward": 1.7138119339942932,
      "reward_std": 0.21042115986347198,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7138119041919708,
      "step": 657
    },
    {
      "completion_length": 104.5625,
      "epoch": 0.42424242424242425,
      "grad_norm": 14.612245559692383,
      "kl": 0.093994140625,
      "learning_rate": 7.878787878787878e-07,
      "loss": 0.0038,
      "reward": 1.4366269707679749,
      "reward_std": 0.07526460662484169,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43662700057029724,
      "step": 658
    },
    {
      "completion_length": 100.453125,
      "epoch": 0.42488716956802064,
      "grad_norm": 50.96489715576172,
      "kl": 0.1083984375,
      "learning_rate": 7.875564152159897e-07,
      "loss": 0.0043,
      "reward": 1.4827229976654053,
      "reward_std": 0.164472047239542,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4827229678630829,
      "step": 659
    },
    {
      "completion_length": 104.296875,
      "epoch": 0.425531914893617,
      "grad_norm": 55.485740661621094,
      "kl": 0.16552734375,
      "learning_rate": 7.872340425531915e-07,
      "loss": 0.0066,
      "reward": 1.5874009132385254,
      "reward_std": 0.17585615068674088,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5874009132385254,
      "step": 660
    },
    {
      "completion_length": 89.46875,
      "epoch": 0.4261766602192134,
      "grad_norm": 10.867607116699219,
      "kl": 0.119140625,
      "learning_rate": 7.869116698903932e-07,
      "loss": 0.0048,
      "reward": 1.6328694224357605,
      "reward_std": 0.12034214287996292,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6328694224357605,
      "step": 661
    },
    {
      "completion_length": 101.515625,
      "epoch": 0.4268214055448098,
      "grad_norm": 44.72108840942383,
      "kl": 0.12548828125,
      "learning_rate": 7.86589297227595e-07,
      "loss": 0.005,
      "reward": 1.4442914724349976,
      "reward_std": 0.12755419686436653,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44429148733615875,
      "step": 662
    },
    {
      "completion_length": 104.09375,
      "epoch": 0.4274661508704062,
      "grad_norm": 19.304611206054688,
      "kl": 0.1318359375,
      "learning_rate": 7.862669245647968e-07,
      "loss": 0.0053,
      "reward": 1.5890682339668274,
      "reward_std": 0.15507179126143456,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5890682637691498,
      "step": 663
    },
    {
      "completion_length": 81.5,
      "epoch": 0.42811089619600257,
      "grad_norm": 14.729875564575195,
      "kl": 0.1669921875,
      "learning_rate": 7.859445519019987e-07,
      "loss": 0.0067,
      "reward": 1.5029385685920715,
      "reward_std": 0.19087035208940506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5029385983943939,
      "step": 664
    },
    {
      "completion_length": 95.34375,
      "epoch": 0.42875564152159895,
      "grad_norm": 15.593925476074219,
      "kl": 0.12890625,
      "learning_rate": 7.856221792392004e-07,
      "loss": 0.0052,
      "reward": 1.4721840620040894,
      "reward_std": 0.1463918462395668,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47218403220176697,
      "step": 665
    },
    {
      "completion_length": 106.125,
      "epoch": 0.42940038684719534,
      "grad_norm": 22.686920166015625,
      "kl": 0.12060546875,
      "learning_rate": 7.852998065764023e-07,
      "loss": 0.0048,
      "reward": 1.6089572310447693,
      "reward_std": 0.1527450978755951,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6089572459459305,
      "step": 666
    },
    {
      "completion_length": 95.109375,
      "epoch": 0.4300451321727917,
      "grad_norm": 10.141805648803711,
      "kl": 0.11474609375,
      "learning_rate": 7.84977433913604e-07,
      "loss": 0.0046,
      "reward": 1.532123863697052,
      "reward_std": 0.12704012170433998,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5321238338947296,
      "step": 667
    },
    {
      "completion_length": 95.859375,
      "epoch": 0.43068987749838816,
      "grad_norm": 9.21531867980957,
      "kl": 0.1357421875,
      "learning_rate": 7.846550612508059e-07,
      "loss": 0.0054,
      "reward": 1.5557440519332886,
      "reward_std": 0.15990904718637466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5557440519332886,
      "step": 668
    },
    {
      "completion_length": 91.984375,
      "epoch": 0.43133462282398455,
      "grad_norm": 15.773168563842773,
      "kl": 0.134033203125,
      "learning_rate": 7.843326885880077e-07,
      "loss": 0.0054,
      "reward": 1.729407250881195,
      "reward_std": 0.11419141665101051,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7294073700904846,
      "step": 669
    },
    {
      "completion_length": 95.765625,
      "epoch": 0.43197936814958093,
      "grad_norm": 17.692100524902344,
      "kl": 0.114013671875,
      "learning_rate": 7.840103159252095e-07,
      "loss": 0.0046,
      "reward": 1.602592945098877,
      "reward_std": 0.1361280083656311,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6025929152965546,
      "step": 670
    },
    {
      "completion_length": 86.0,
      "epoch": 0.4326241134751773,
      "grad_norm": 18.247699737548828,
      "kl": 0.12939453125,
      "learning_rate": 7.836879432624112e-07,
      "loss": 0.0052,
      "reward": 1.5443543195724487,
      "reward_std": 0.10458062216639519,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5443542748689651,
      "step": 671
    },
    {
      "completion_length": 97.546875,
      "epoch": 0.4332688588007737,
      "grad_norm": 22.875259399414062,
      "kl": 0.1044921875,
      "learning_rate": 7.833655705996132e-07,
      "loss": 0.0042,
      "reward": 1.4106297492980957,
      "reward_std": 0.07918915897607803,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4106297940015793,
      "step": 672
    },
    {
      "completion_length": 96.734375,
      "epoch": 0.4339136041263701,
      "grad_norm": 16.27412223815918,
      "kl": 0.1396484375,
      "learning_rate": 7.830431979368149e-07,
      "loss": 0.0056,
      "reward": 1.5767380595207214,
      "reward_std": 0.09661797434091568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5767380446195602,
      "step": 673
    },
    {
      "completion_length": 99.25,
      "epoch": 0.4345583494519665,
      "grad_norm": 216.69961547851562,
      "kl": 0.13134765625,
      "learning_rate": 7.827208252740167e-07,
      "loss": 0.0052,
      "reward": 1.5629220008850098,
      "reward_std": 0.11474324017763138,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5629220306873322,
      "step": 674
    },
    {
      "completion_length": 99.34375,
      "epoch": 0.43520309477756286,
      "grad_norm": 11.57163143157959,
      "kl": 0.118408203125,
      "learning_rate": 7.823984526112185e-07,
      "loss": 0.0047,
      "reward": 1.3828779458999634,
      "reward_std": 0.08658572658896446,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.38287800550460815,
      "step": 675
    },
    {
      "completion_length": 94.59375,
      "epoch": 0.43584784010315925,
      "grad_norm": 31.561555862426758,
      "kl": 0.115966796875,
      "learning_rate": 7.820760799484203e-07,
      "loss": 0.0046,
      "reward": 1.5925236344337463,
      "reward_std": 0.10616682469844818,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5925236642360687,
      "step": 676
    },
    {
      "completion_length": 95.0625,
      "epoch": 0.43649258542875563,
      "grad_norm": 8.750934600830078,
      "kl": 0.109130859375,
      "learning_rate": 7.817537072856221e-07,
      "loss": 0.0044,
      "reward": 1.5274120569229126,
      "reward_std": 0.15606505796313286,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.527412086725235,
      "step": 677
    },
    {
      "completion_length": 101.875,
      "epoch": 0.437137330754352,
      "grad_norm": 144.177978515625,
      "kl": 0.116943359375,
      "learning_rate": 7.81431334622824e-07,
      "loss": 0.0047,
      "reward": 1.5111930966377258,
      "reward_std": 0.1961452066898346,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5268180966377258,
      "step": 678
    },
    {
      "completion_length": 87.703125,
      "epoch": 0.4377820760799484,
      "grad_norm": 13.244322776794434,
      "kl": 0.123779296875,
      "learning_rate": 7.811089619600257e-07,
      "loss": 0.005,
      "reward": 1.52597177028656,
      "reward_std": 0.12621500715613365,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5259717255830765,
      "step": 679
    },
    {
      "completion_length": 102.34375,
      "epoch": 0.4384268214055448,
      "grad_norm": 11.01314926147461,
      "kl": 0.11767578125,
      "learning_rate": 7.807865892972275e-07,
      "loss": 0.0047,
      "reward": 1.4737398028373718,
      "reward_std": 0.10230442509055138,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4737398326396942,
      "step": 680
    },
    {
      "completion_length": 90.6875,
      "epoch": 0.43907156673114117,
      "grad_norm": 13.169625282287598,
      "kl": 0.13720703125,
      "learning_rate": 7.804642166344294e-07,
      "loss": 0.0055,
      "reward": 1.5600587129592896,
      "reward_std": 0.13433203101158142,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5756836533546448,
      "step": 681
    },
    {
      "completion_length": 96.671875,
      "epoch": 0.4397163120567376,
      "grad_norm": 19.183500289916992,
      "kl": 0.1298828125,
      "learning_rate": 7.801418439716312e-07,
      "loss": 0.0052,
      "reward": 1.52139014005661,
      "reward_std": 0.1388275846838951,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5213901400566101,
      "step": 682
    },
    {
      "completion_length": 119.015625,
      "epoch": 0.440361057382334,
      "grad_norm": 9.817489624023438,
      "kl": 0.103515625,
      "learning_rate": 7.798194713088329e-07,
      "loss": 0.0041,
      "reward": 1.6039352416992188,
      "reward_std": 0.0923042930662632,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6039351522922516,
      "step": 683
    },
    {
      "completion_length": 92.75,
      "epoch": 0.4410058027079304,
      "grad_norm": 17.6253662109375,
      "kl": 0.1904296875,
      "learning_rate": 7.794970986460347e-07,
      "loss": 0.0076,
      "reward": 1.553667962551117,
      "reward_std": 0.14492272213101387,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5536679923534393,
      "step": 684
    },
    {
      "completion_length": 97.59375,
      "epoch": 0.44165054803352677,
      "grad_norm": 21.793113708496094,
      "kl": 0.13134765625,
      "learning_rate": 7.791747259832367e-07,
      "loss": 0.0052,
      "reward": 1.608609676361084,
      "reward_std": 0.1321398764848709,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6086097061634064,
      "step": 685
    },
    {
      "completion_length": 91.28125,
      "epoch": 0.44229529335912315,
      "grad_norm": 15.338560104370117,
      "kl": 0.12939453125,
      "learning_rate": 7.788523533204384e-07,
      "loss": 0.0052,
      "reward": 1.4791777729988098,
      "reward_std": 0.06188085302710533,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4791778028011322,
      "step": 686
    },
    {
      "completion_length": 95.234375,
      "epoch": 0.44294003868471954,
      "grad_norm": 56.819026947021484,
      "kl": 0.11328125,
      "learning_rate": 7.785299806576401e-07,
      "loss": 0.0045,
      "reward": 1.608103096485138,
      "reward_std": 0.08347269147634506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6081030368804932,
      "step": 687
    },
    {
      "completion_length": 97.890625,
      "epoch": 0.4435847840103159,
      "grad_norm": 38.48658752441406,
      "kl": 0.129638671875,
      "learning_rate": 7.78207607994842e-07,
      "loss": 0.0052,
      "reward": 1.4730200171470642,
      "reward_std": 0.12290628626942635,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4730200171470642,
      "step": 688
    },
    {
      "completion_length": 99.21875,
      "epoch": 0.4442295293359123,
      "grad_norm": 8.553914070129395,
      "kl": 0.1357421875,
      "learning_rate": 7.778852353320437e-07,
      "loss": 0.0054,
      "reward": 1.4979106187820435,
      "reward_std": 0.07518330216407776,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49791061878204346,
      "step": 689
    },
    {
      "completion_length": 96.625,
      "epoch": 0.4448742746615087,
      "grad_norm": 13.162261009216309,
      "kl": 0.123291015625,
      "learning_rate": 7.775628626692456e-07,
      "loss": 0.0049,
      "reward": 1.5289295315742493,
      "reward_std": 0.13452866673469543,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5289295315742493,
      "step": 690
    },
    {
      "completion_length": 97.515625,
      "epoch": 0.4455190199871051,
      "grad_norm": 33.99656295776367,
      "kl": 0.126953125,
      "learning_rate": 7.772404900064475e-07,
      "loss": 0.0051,
      "reward": 1.5598184466362,
      "reward_std": 0.07671795412898064,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5598184615373611,
      "step": 691
    },
    {
      "completion_length": 112.625,
      "epoch": 0.44616376531270147,
      "grad_norm": 48.81503677368164,
      "kl": 0.1162109375,
      "learning_rate": 7.769181173436492e-07,
      "loss": 0.0046,
      "reward": 1.4112409949302673,
      "reward_std": 0.18779970705509186,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.42686596512794495,
      "step": 692
    },
    {
      "completion_length": 97.609375,
      "epoch": 0.44680851063829785,
      "grad_norm": 26.25737762451172,
      "kl": 0.13232421875,
      "learning_rate": 7.765957446808509e-07,
      "loss": 0.0053,
      "reward": 1.6583746671676636,
      "reward_std": 0.12108327448368073,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6583746075630188,
      "step": 693
    },
    {
      "completion_length": 100.453125,
      "epoch": 0.44745325596389424,
      "grad_norm": 15.020151138305664,
      "kl": 0.10986328125,
      "learning_rate": 7.762733720180529e-07,
      "loss": 0.0044,
      "reward": 1.4644600749015808,
      "reward_std": 0.13533856347203255,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4644601047039032,
      "step": 694
    },
    {
      "completion_length": 95.71875,
      "epoch": 0.4480980012894907,
      "grad_norm": 21.066251754760742,
      "kl": 0.106689453125,
      "learning_rate": 7.759509993552547e-07,
      "loss": 0.0043,
      "reward": 1.6312799453735352,
      "reward_std": 0.10122665017843246,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6312799155712128,
      "step": 695
    },
    {
      "completion_length": 103.28125,
      "epoch": 0.44874274661508706,
      "grad_norm": 11.778822898864746,
      "kl": 0.11962890625,
      "learning_rate": 7.756286266924564e-07,
      "loss": 0.0048,
      "reward": 1.6831336617469788,
      "reward_std": 0.06576647609472275,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6831336915493011,
      "step": 696
    },
    {
      "completion_length": 97.890625,
      "epoch": 0.44938749194068345,
      "grad_norm": 24.91705894470215,
      "kl": 0.12060546875,
      "learning_rate": 7.753062540296583e-07,
      "loss": 0.0048,
      "reward": 1.7192001342773438,
      "reward_std": 0.0967547558248043,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7192001640796661,
      "step": 697
    },
    {
      "completion_length": 93.796875,
      "epoch": 0.45003223726627983,
      "grad_norm": 10.502815246582031,
      "kl": 0.138671875,
      "learning_rate": 7.749838813668601e-07,
      "loss": 0.0055,
      "reward": 1.5451446175575256,
      "reward_std": 0.10906476899981499,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5451445579528809,
      "step": 698
    },
    {
      "completion_length": 91.296875,
      "epoch": 0.4506769825918762,
      "grad_norm": 19.44222068786621,
      "kl": 0.1357421875,
      "learning_rate": 7.746615087040619e-07,
      "loss": 0.0054,
      "reward": 1.5762072801589966,
      "reward_std": 0.09864221513271332,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5762072950601578,
      "step": 699
    },
    {
      "completion_length": 97.125,
      "epoch": 0.4513217279174726,
      "grad_norm": 22.833763122558594,
      "kl": 0.129150390625,
      "learning_rate": 7.743391360412637e-07,
      "loss": 0.0052,
      "reward": 1.4687912464141846,
      "reward_std": 0.09121186658740044,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4687913507223129,
      "step": 700
    },
    {
      "completion_length": 91.734375,
      "epoch": 0.451966473243069,
      "grad_norm": 30.23716163635254,
      "kl": 0.1455078125,
      "learning_rate": 7.740167633784655e-07,
      "loss": 0.0058,
      "reward": 1.6057734489440918,
      "reward_std": 0.0660810861736536,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6057734489440918,
      "step": 701
    },
    {
      "completion_length": 88.265625,
      "epoch": 0.4526112185686654,
      "grad_norm": 15.325862884521484,
      "kl": 0.11572265625,
      "learning_rate": 7.736943907156672e-07,
      "loss": 0.0046,
      "reward": 1.4314576387405396,
      "reward_std": 0.08893175795674324,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43145765364170074,
      "step": 702
    },
    {
      "completion_length": 95.75,
      "epoch": 0.45325596389426176,
      "grad_norm": 12.580069541931152,
      "kl": 0.13623046875,
      "learning_rate": 7.733720180528692e-07,
      "loss": 0.0054,
      "reward": 1.5887573957443237,
      "reward_std": 0.07212099991738796,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5887573659420013,
      "step": 703
    },
    {
      "completion_length": 92.375,
      "epoch": 0.45390070921985815,
      "grad_norm": 16.807231903076172,
      "kl": 0.109375,
      "learning_rate": 7.730496453900709e-07,
      "loss": 0.0044,
      "reward": 1.5106210708618164,
      "reward_std": 0.08820386417210102,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5106211006641388,
      "step": 704
    },
    {
      "completion_length": 93.21875,
      "epoch": 0.45454545454545453,
      "grad_norm": 519.1197509765625,
      "kl": 0.106689453125,
      "learning_rate": 7.727272727272727e-07,
      "loss": 0.0043,
      "reward": 1.5637730956077576,
      "reward_std": 0.10183101892471313,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.56377312541008,
      "step": 705
    },
    {
      "completion_length": 83.203125,
      "epoch": 0.4551901998710509,
      "grad_norm": 46.5788688659668,
      "kl": 0.15185546875,
      "learning_rate": 7.724049000644744e-07,
      "loss": 0.0061,
      "reward": 1.6154998540878296,
      "reward_std": 0.10929432883858681,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6154997944831848,
      "step": 706
    },
    {
      "completion_length": 86.953125,
      "epoch": 0.4558349451966473,
      "grad_norm": 15.66248893737793,
      "kl": 0.133544921875,
      "learning_rate": 7.720825274016764e-07,
      "loss": 0.0053,
      "reward": 1.4980468153953552,
      "reward_std": 0.061294110491871834,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49804680049419403,
      "step": 707
    },
    {
      "completion_length": 102.515625,
      "epoch": 0.4564796905222437,
      "grad_norm": 16.94169807434082,
      "kl": 0.1357421875,
      "learning_rate": 7.717601547388781e-07,
      "loss": 0.0054,
      "reward": 1.716218113899231,
      "reward_std": 0.09692734852433205,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7162180542945862,
      "step": 708
    },
    {
      "completion_length": 97.265625,
      "epoch": 0.4571244358478401,
      "grad_norm": 10.600395202636719,
      "kl": 0.114501953125,
      "learning_rate": 7.7143778207608e-07,
      "loss": 0.0046,
      "reward": 1.5950488448143005,
      "reward_std": 0.0715242438018322,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5950488448143005,
      "step": 709
    },
    {
      "completion_length": 87.515625,
      "epoch": 0.4577691811734365,
      "grad_norm": 11.393702507019043,
      "kl": 0.1279296875,
      "learning_rate": 7.711154094132817e-07,
      "loss": 0.0051,
      "reward": 1.577049434185028,
      "reward_std": 0.09067998081445694,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5770494043827057,
      "step": 710
    },
    {
      "completion_length": 94.21875,
      "epoch": 0.4584139264990329,
      "grad_norm": 17.261930465698242,
      "kl": 0.111572265625,
      "learning_rate": 7.707930367504836e-07,
      "loss": 0.0045,
      "reward": 1.6038063764572144,
      "reward_std": 0.10224081575870514,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6038063764572144,
      "step": 711
    },
    {
      "completion_length": 101.921875,
      "epoch": 0.4590586718246293,
      "grad_norm": 14.559025764465332,
      "kl": 0.112548828125,
      "learning_rate": 7.704706640876853e-07,
      "loss": 0.0045,
      "reward": 1.4666990041732788,
      "reward_std": 0.11419890448451042,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4666990488767624,
      "step": 712
    },
    {
      "completion_length": 98.171875,
      "epoch": 0.45970341715022567,
      "grad_norm": 15.404255867004395,
      "kl": 0.148681640625,
      "learning_rate": 7.701482914248872e-07,
      "loss": 0.006,
      "reward": 1.6305444836616516,
      "reward_std": 0.11607625335454941,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6305445432662964,
      "step": 713
    },
    {
      "completion_length": 101.125,
      "epoch": 0.46034816247582205,
      "grad_norm": 19.92659568786621,
      "kl": 0.125,
      "learning_rate": 7.698259187620889e-07,
      "loss": 0.005,
      "reward": 1.4447981715202332,
      "reward_std": 0.1305028274655342,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44479817152023315,
      "step": 714
    },
    {
      "completion_length": 100.625,
      "epoch": 0.46099290780141844,
      "grad_norm": 11.078210830688477,
      "kl": 0.1474609375,
      "learning_rate": 7.695035460992907e-07,
      "loss": 0.0059,
      "reward": 1.4863204956054688,
      "reward_std": 0.09369765594601631,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48632049560546875,
      "step": 715
    },
    {
      "completion_length": 94.859375,
      "epoch": 0.4616376531270148,
      "grad_norm": 16.100868225097656,
      "kl": 0.13134765625,
      "learning_rate": 7.691811734364926e-07,
      "loss": 0.0053,
      "reward": 1.611817181110382,
      "reward_std": 0.09366685152053833,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6118170917034149,
      "step": 716
    },
    {
      "completion_length": 101.265625,
      "epoch": 0.4622823984526112,
      "grad_norm": 21.628129959106445,
      "kl": 0.110595703125,
      "learning_rate": 7.688588007736944e-07,
      "loss": 0.0044,
      "reward": 1.6645547151565552,
      "reward_std": 0.1055876649916172,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6645546853542328,
      "step": 717
    },
    {
      "completion_length": 102.21875,
      "epoch": 0.4629271437782076,
      "grad_norm": 20.826440811157227,
      "kl": 0.132568359375,
      "learning_rate": 7.685364281108961e-07,
      "loss": 0.0053,
      "reward": 1.5520586371421814,
      "reward_std": 0.09113192185759544,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5520585924386978,
      "step": 718
    },
    {
      "completion_length": 99.46875,
      "epoch": 0.463571889103804,
      "grad_norm": 16.853395462036133,
      "kl": 0.13525390625,
      "learning_rate": 7.68214055448098e-07,
      "loss": 0.0054,
      "reward": 1.5113345980644226,
      "reward_std": 0.11239676922559738,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5113345384597778,
      "step": 719
    },
    {
      "completion_length": 103.890625,
      "epoch": 0.46421663442940037,
      "grad_norm": 12.532203674316406,
      "kl": 0.13525390625,
      "learning_rate": 7.678916827852998e-07,
      "loss": 0.0054,
      "reward": 1.4734601974487305,
      "reward_std": 0.06641046330332756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47346019744873047,
      "step": 720
    },
    {
      "completion_length": 110.796875,
      "epoch": 0.46486137975499675,
      "grad_norm": 10.890167236328125,
      "kl": 0.12841796875,
      "learning_rate": 7.675693101225016e-07,
      "loss": 0.0051,
      "reward": 1.4852750301361084,
      "reward_std": 0.061814336106181145,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4852750599384308,
      "step": 721
    },
    {
      "completion_length": 116.859375,
      "epoch": 0.4655061250805932,
      "grad_norm": 10.753568649291992,
      "kl": 0.1123046875,
      "learning_rate": 7.672469374597034e-07,
      "loss": 0.0045,
      "reward": 1.4886300563812256,
      "reward_std": 0.08008428663015366,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4886299967765808,
      "step": 722
    },
    {
      "completion_length": 109.625,
      "epoch": 0.4661508704061896,
      "grad_norm": 10.12906551361084,
      "kl": 0.12255859375,
      "learning_rate": 7.669245647969052e-07,
      "loss": 0.0049,
      "reward": 1.503622591495514,
      "reward_std": 0.07655685395002365,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5036226361989975,
      "step": 723
    },
    {
      "completion_length": 118.53125,
      "epoch": 0.46679561573178596,
      "grad_norm": 17.245555877685547,
      "kl": 0.126220703125,
      "learning_rate": 7.66602192134107e-07,
      "loss": 0.0051,
      "reward": 1.6156468391418457,
      "reward_std": 0.11876443773508072,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6156468689441681,
      "step": 724
    },
    {
      "completion_length": 125.328125,
      "epoch": 0.46744036105738235,
      "grad_norm": 7.821877956390381,
      "kl": 0.1220703125,
      "learning_rate": 7.662798194713089e-07,
      "loss": 0.0049,
      "reward": 1.6727633476257324,
      "reward_std": 0.12070934474468231,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6727633774280548,
      "step": 725
    },
    {
      "completion_length": 127.796875,
      "epoch": 0.46808510638297873,
      "grad_norm": 18.543737411499023,
      "kl": 0.131591796875,
      "learning_rate": 7.659574468085106e-07,
      "loss": 0.0053,
      "reward": 1.5528809428215027,
      "reward_std": 0.09312764927744865,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5528809428215027,
      "step": 726
    },
    {
      "completion_length": 117.171875,
      "epoch": 0.4687298517085751,
      "grad_norm": 21.947412490844727,
      "kl": 0.117919921875,
      "learning_rate": 7.656350741457124e-07,
      "loss": 0.0047,
      "reward": 1.6618314385414124,
      "reward_std": 0.0795222856104374,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6618314981460571,
      "step": 727
    },
    {
      "completion_length": 110.140625,
      "epoch": 0.4693745970341715,
      "grad_norm": 10.362268447875977,
      "kl": 0.138671875,
      "learning_rate": 7.653127014829143e-07,
      "loss": 0.0056,
      "reward": 1.5495651364326477,
      "reward_std": 0.0900680385529995,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5495651364326477,
      "step": 728
    },
    {
      "completion_length": 122.140625,
      "epoch": 0.4700193423597679,
      "grad_norm": 9.733297348022461,
      "kl": 0.12158203125,
      "learning_rate": 7.649903288201161e-07,
      "loss": 0.0049,
      "reward": 1.6779837608337402,
      "reward_std": 0.06389831379055977,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6779836416244507,
      "step": 729
    },
    {
      "completion_length": 121.40625,
      "epoch": 0.4706640876853643,
      "grad_norm": 153.9310760498047,
      "kl": 0.1064453125,
      "learning_rate": 7.646679561573178e-07,
      "loss": 0.0043,
      "reward": 1.5564913749694824,
      "reward_std": 0.18158581107854843,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5564914643764496,
      "step": 730
    },
    {
      "completion_length": 118.234375,
      "epoch": 0.47130883301096066,
      "grad_norm": 9.935221672058105,
      "kl": 0.123779296875,
      "learning_rate": 7.643455834945196e-07,
      "loss": 0.0049,
      "reward": 1.5225478410720825,
      "reward_std": 0.09475431591272354,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5225478410720825,
      "step": 731
    },
    {
      "completion_length": 124.953125,
      "epoch": 0.47195357833655704,
      "grad_norm": 20.191246032714844,
      "kl": 0.1259765625,
      "learning_rate": 7.640232108317214e-07,
      "loss": 0.0051,
      "reward": 1.552919089794159,
      "reward_std": 0.07817603275179863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5529190897941589,
      "step": 732
    },
    {
      "completion_length": 110.5625,
      "epoch": 0.47259832366215343,
      "grad_norm": 18.04339027404785,
      "kl": 0.11865234375,
      "learning_rate": 7.637008381689233e-07,
      "loss": 0.0047,
      "reward": 1.5138545632362366,
      "reward_std": 0.13918767496943474,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5138545781373978,
      "step": 733
    },
    {
      "completion_length": 114.0,
      "epoch": 0.4732430689877498,
      "grad_norm": 10.329635620117188,
      "kl": 0.11767578125,
      "learning_rate": 7.63378465506125e-07,
      "loss": 0.0047,
      "reward": 1.6484586000442505,
      "reward_std": 0.11795339360833168,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6484585404396057,
      "step": 734
    },
    {
      "completion_length": 108.28125,
      "epoch": 0.4738878143133462,
      "grad_norm": 19.822538375854492,
      "kl": 0.128662109375,
      "learning_rate": 7.630560928433269e-07,
      "loss": 0.0051,
      "reward": 1.5399014949798584,
      "reward_std": 0.11883477121591568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.53990139067173,
      "step": 735
    },
    {
      "completion_length": 115.875,
      "epoch": 0.47453255963894264,
      "grad_norm": 31.11959457397461,
      "kl": 0.134765625,
      "learning_rate": 7.627337201805286e-07,
      "loss": 0.0054,
      "reward": 1.5331445336341858,
      "reward_std": 0.15693273022770882,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5487695634365082,
      "step": 736
    },
    {
      "completion_length": 106.453125,
      "epoch": 0.475177304964539,
      "grad_norm": 14.009469985961914,
      "kl": 0.1240234375,
      "learning_rate": 7.624113475177305e-07,
      "loss": 0.005,
      "reward": 1.4909018278121948,
      "reward_std": 0.11989344283938408,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4909018278121948,
      "step": 737
    },
    {
      "completion_length": 120.765625,
      "epoch": 0.4758220502901354,
      "grad_norm": 13.631786346435547,
      "kl": 0.1240234375,
      "learning_rate": 7.620889748549323e-07,
      "loss": 0.005,
      "reward": 1.5898830890655518,
      "reward_std": 0.08887490257620811,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.589883029460907,
      "step": 738
    },
    {
      "completion_length": 105.171875,
      "epoch": 0.4764667956157318,
      "grad_norm": 13.331558227539062,
      "kl": 0.12744140625,
      "learning_rate": 7.617666021921341e-07,
      "loss": 0.0051,
      "reward": 1.5238662362098694,
      "reward_std": 0.08402681723237038,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5238662660121918,
      "step": 739
    },
    {
      "completion_length": 107.03125,
      "epoch": 0.4771115409413282,
      "grad_norm": 25.664894104003906,
      "kl": 0.116943359375,
      "learning_rate": 7.614442295293358e-07,
      "loss": 0.0047,
      "reward": 1.5047930479049683,
      "reward_std": 0.11811169236898422,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5047930777072906,
      "step": 740
    },
    {
      "completion_length": 104.046875,
      "epoch": 0.47775628626692457,
      "grad_norm": 7.534611701965332,
      "kl": 0.154296875,
      "learning_rate": 7.611218568665378e-07,
      "loss": 0.0062,
      "reward": 1.4220759868621826,
      "reward_std": 0.11570665240287781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4220760092139244,
      "step": 741
    },
    {
      "completion_length": 104.78125,
      "epoch": 0.47840103159252095,
      "grad_norm": 96.45923614501953,
      "kl": 0.118408203125,
      "learning_rate": 7.607994842037395e-07,
      "loss": 0.0047,
      "reward": 1.4136383533477783,
      "reward_std": 0.09354563057422638,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41363826394081116,
      "step": 742
    },
    {
      "completion_length": 113.328125,
      "epoch": 0.47904577691811734,
      "grad_norm": 29.232112884521484,
      "kl": 0.14111328125,
      "learning_rate": 7.604771115409413e-07,
      "loss": 0.0056,
      "reward": 1.5308947563171387,
      "reward_std": 0.11985908448696136,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5308947861194611,
      "step": 743
    },
    {
      "completion_length": 99.03125,
      "epoch": 0.4796905222437137,
      "grad_norm": 22.255399703979492,
      "kl": 0.13720703125,
      "learning_rate": 7.601547388781431e-07,
      "loss": 0.0055,
      "reward": 1.6334994435310364,
      "reward_std": 0.10051496885716915,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6334995031356812,
      "step": 744
    },
    {
      "completion_length": 87.390625,
      "epoch": 0.4803352675693101,
      "grad_norm": 16.31736946105957,
      "kl": 0.2001953125,
      "learning_rate": 7.598323662153449e-07,
      "loss": 0.008,
      "reward": 1.5974202156066895,
      "reward_std": 0.09732247516512871,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5974202156066895,
      "step": 745
    },
    {
      "completion_length": 110.53125,
      "epoch": 0.4809800128949065,
      "grad_norm": 8.461565971374512,
      "kl": 0.13525390625,
      "learning_rate": 7.595099935525467e-07,
      "loss": 0.0054,
      "reward": 1.7255396246910095,
      "reward_std": 0.1103828065097332,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7255396246910095,
      "step": 746
    },
    {
      "completion_length": 96.4375,
      "epoch": 0.4816247582205029,
      "grad_norm": 225.49200439453125,
      "kl": 0.126953125,
      "learning_rate": 7.591876208897486e-07,
      "loss": 0.0051,
      "reward": 1.5281953811645508,
      "reward_std": 0.08646541833877563,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5281953811645508,
      "step": 747
    },
    {
      "completion_length": 92.984375,
      "epoch": 0.48226950354609927,
      "grad_norm": 89.93632507324219,
      "kl": 0.13916015625,
      "learning_rate": 7.588652482269503e-07,
      "loss": 0.0056,
      "reward": 1.652679681777954,
      "reward_std": 0.11553067341446877,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6526796817779541,
      "step": 748
    },
    {
      "completion_length": 104.0,
      "epoch": 0.4829142488716957,
      "grad_norm": 9.606204986572266,
      "kl": 0.136962890625,
      "learning_rate": 7.585428755641521e-07,
      "loss": 0.0055,
      "reward": 1.4608428478240967,
      "reward_std": 0.101207185536623,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4608428180217743,
      "step": 749
    },
    {
      "completion_length": 80.453125,
      "epoch": 0.4835589941972921,
      "grad_norm": 11.411137580871582,
      "kl": 0.14111328125,
      "learning_rate": 7.58220502901354e-07,
      "loss": 0.0056,
      "reward": 1.5915326476097107,
      "reward_std": 0.13187076896429062,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5915326774120331,
      "step": 750
    },
    {
      "completion_length": 98.859375,
      "epoch": 0.4842037395228885,
      "grad_norm": 43.71159362792969,
      "kl": 0.14501953125,
      "learning_rate": 7.578981302385558e-07,
      "loss": 0.0058,
      "reward": 1.5974067449569702,
      "reward_std": 0.13504815101623535,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5974067747592926,
      "step": 751
    },
    {
      "completion_length": 100.25,
      "epoch": 0.48484848484848486,
      "grad_norm": 10.606791496276855,
      "kl": 0.12939453125,
      "learning_rate": 7.575757575757575e-07,
      "loss": 0.0052,
      "reward": 1.5629411935806274,
      "reward_std": 0.05437769740819931,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5629411041736603,
      "step": 752
    },
    {
      "completion_length": 93.09375,
      "epoch": 0.48549323017408125,
      "grad_norm": 25.361251831054688,
      "kl": 0.1533203125,
      "learning_rate": 7.572533849129593e-07,
      "loss": 0.0061,
      "reward": 1.5369932055473328,
      "reward_std": 0.10840344429016113,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5369932353496552,
      "step": 753
    },
    {
      "completion_length": 96.796875,
      "epoch": 0.48613797549967763,
      "grad_norm": 35.501617431640625,
      "kl": 0.1416015625,
      "learning_rate": 7.569310122501612e-07,
      "loss": 0.0057,
      "reward": 1.5901548862457275,
      "reward_std": 0.122325100004673,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5901548862457275,
      "step": 754
    },
    {
      "completion_length": 98.59375,
      "epoch": 0.486782720825274,
      "grad_norm": 45.77256774902344,
      "kl": 0.220703125,
      "learning_rate": 7.56608639587363e-07,
      "loss": 0.0088,
      "reward": 1.449923574924469,
      "reward_std": 0.12096518278121948,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.449923574924469,
      "step": 755
    },
    {
      "completion_length": 100.0,
      "epoch": 0.4874274661508704,
      "grad_norm": 14.880589485168457,
      "kl": 0.1396484375,
      "learning_rate": 7.562862669245647e-07,
      "loss": 0.0056,
      "reward": 1.5938887000083923,
      "reward_std": 0.16511598601937294,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6095137000083923,
      "step": 756
    },
    {
      "completion_length": 111.859375,
      "epoch": 0.4880722114764668,
      "grad_norm": 10.331930160522461,
      "kl": 0.147705078125,
      "learning_rate": 7.559638942617666e-07,
      "loss": 0.0059,
      "reward": 1.5344634056091309,
      "reward_std": 0.1292192693799734,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5344634652137756,
      "step": 757
    },
    {
      "completion_length": 107.359375,
      "epoch": 0.4887169568020632,
      "grad_norm": 11.895607948303223,
      "kl": 0.1123046875,
      "learning_rate": 7.556415215989683e-07,
      "loss": 0.0045,
      "reward": 1.4465181231498718,
      "reward_std": 0.10399571061134338,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4621431678533554,
      "step": 758
    },
    {
      "completion_length": 102.390625,
      "epoch": 0.48936170212765956,
      "grad_norm": 28.140899658203125,
      "kl": 0.118408203125,
      "learning_rate": 7.553191489361702e-07,
      "loss": 0.0047,
      "reward": 1.4872334003448486,
      "reward_std": 0.1054934673011303,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48723337054252625,
      "step": 759
    },
    {
      "completion_length": 98.296875,
      "epoch": 0.49000644745325594,
      "grad_norm": 8.313096046447754,
      "kl": 0.137939453125,
      "learning_rate": 7.54996776273372e-07,
      "loss": 0.0055,
      "reward": 1.4911122918128967,
      "reward_std": 0.06940718367695808,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49111229181289673,
      "step": 760
    },
    {
      "completion_length": 100.96875,
      "epoch": 0.49065119277885233,
      "grad_norm": 25.631065368652344,
      "kl": 0.1279296875,
      "learning_rate": 7.546744036105738e-07,
      "loss": 0.0051,
      "reward": 1.6772446632385254,
      "reward_std": 0.11209296062588692,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.677244633436203,
      "step": 761
    },
    {
      "completion_length": 98.25,
      "epoch": 0.4912959381044487,
      "grad_norm": 15.898694038391113,
      "kl": 0.14208984375,
      "learning_rate": 7.543520309477755e-07,
      "loss": 0.0057,
      "reward": 1.5463205575942993,
      "reward_std": 0.10821164213120937,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5463205575942993,
      "step": 762
    },
    {
      "completion_length": 106.59375,
      "epoch": 0.49194068343004516,
      "grad_norm": 14.443410873413086,
      "kl": 0.11767578125,
      "learning_rate": 7.540296582849775e-07,
      "loss": 0.0047,
      "reward": 1.4314526915550232,
      "reward_std": 0.09445375949144363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4314526617527008,
      "step": 763
    },
    {
      "completion_length": 106.140625,
      "epoch": 0.49258542875564154,
      "grad_norm": 34.89936447143555,
      "kl": 0.11767578125,
      "learning_rate": 7.537072856221792e-07,
      "loss": 0.0047,
      "reward": 1.5694019794464111,
      "reward_std": 0.08024864830076694,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5694020092487335,
      "step": 764
    },
    {
      "completion_length": 103.78125,
      "epoch": 0.4932301740812379,
      "grad_norm": 31.551544189453125,
      "kl": 0.18115234375,
      "learning_rate": 7.53384912959381e-07,
      "loss": 0.0073,
      "reward": 1.4196154475212097,
      "reward_std": 0.10790861025452614,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4196154773235321,
      "step": 765
    },
    {
      "completion_length": 102.65625,
      "epoch": 0.4938749194068343,
      "grad_norm": 8.728348731994629,
      "kl": 0.1455078125,
      "learning_rate": 7.530625402965828e-07,
      "loss": 0.0058,
      "reward": 1.3880661725997925,
      "reward_std": 0.12530139833688736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3880661725997925,
      "step": 766
    },
    {
      "completion_length": 104.390625,
      "epoch": 0.4945196647324307,
      "grad_norm": 34.32732009887695,
      "kl": 0.1494140625,
      "learning_rate": 7.527401676337847e-07,
      "loss": 0.006,
      "reward": 1.3845961093902588,
      "reward_std": 0.1502724587917328,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4002210944890976,
      "step": 767
    },
    {
      "completion_length": 96.828125,
      "epoch": 0.4951644100580271,
      "grad_norm": 12.261150360107422,
      "kl": 0.105712890625,
      "learning_rate": 7.524177949709864e-07,
      "loss": 0.0042,
      "reward": 1.5911886096000671,
      "reward_std": 0.06629323586821556,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5911886692047119,
      "step": 768
    },
    {
      "completion_length": 107.125,
      "epoch": 0.49580915538362347,
      "grad_norm": 14.596806526184082,
      "kl": 0.1103515625,
      "learning_rate": 7.520954223081883e-07,
      "loss": 0.0044,
      "reward": 1.6019713878631592,
      "reward_std": 0.05874236673116684,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6019714027643204,
      "step": 769
    },
    {
      "completion_length": 105.0,
      "epoch": 0.49645390070921985,
      "grad_norm": 15.82909870147705,
      "kl": 0.1240234375,
      "learning_rate": 7.5177304964539e-07,
      "loss": 0.005,
      "reward": 1.3426456451416016,
      "reward_std": 0.06746992282569408,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.34264568984508514,
      "step": 770
    },
    {
      "completion_length": 103.828125,
      "epoch": 0.49709864603481624,
      "grad_norm": 24.44872283935547,
      "kl": 0.1689453125,
      "learning_rate": 7.514506769825918e-07,
      "loss": 0.0067,
      "reward": 1.5089261531829834,
      "reward_std": 0.101194828748703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5089261829853058,
      "step": 771
    },
    {
      "completion_length": 113.609375,
      "epoch": 0.4977433913604126,
      "grad_norm": 10.225950241088867,
      "kl": 0.12744140625,
      "learning_rate": 7.511283043197937e-07,
      "loss": 0.0051,
      "reward": 1.636949062347412,
      "reward_std": 0.09513459354639053,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6369490921497345,
      "step": 772
    },
    {
      "completion_length": 104.296875,
      "epoch": 0.498388136686009,
      "grad_norm": 21.62342643737793,
      "kl": 0.1416015625,
      "learning_rate": 7.508059316569955e-07,
      "loss": 0.0057,
      "reward": 1.6637132167816162,
      "reward_std": 0.11707671359181404,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6637131869792938,
      "step": 773
    },
    {
      "completion_length": 98.796875,
      "epoch": 0.4990328820116054,
      "grad_norm": 56.46780014038086,
      "kl": 0.112060546875,
      "learning_rate": 7.504835589941972e-07,
      "loss": 0.0045,
      "reward": 1.534183382987976,
      "reward_std": 0.09461552277207375,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5341833829879761,
      "step": 774
    },
    {
      "completion_length": 98.484375,
      "epoch": 0.4996776273372018,
      "grad_norm": 12.098061561584473,
      "kl": 0.11767578125,
      "learning_rate": 7.50161186331399e-07,
      "loss": 0.0047,
      "reward": 1.6135407090187073,
      "reward_std": 0.11286146193742752,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6135406941175461,
      "step": 775
    },
    {
      "completion_length": 114.984375,
      "epoch": 0.5003223726627982,
      "grad_norm": 20.275917053222656,
      "kl": 0.10302734375,
      "learning_rate": 7.498388136686009e-07,
      "loss": 0.0041,
      "reward": 1.5216720700263977,
      "reward_std": 0.06171071156859398,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5216720998287201,
      "step": 776
    },
    {
      "completion_length": 104.53125,
      "epoch": 0.5009671179883946,
      "grad_norm": 13.640120506286621,
      "kl": 0.1298828125,
      "learning_rate": 7.495164410058027e-07,
      "loss": 0.0052,
      "reward": 1.6737434267997742,
      "reward_std": 0.11507437378168106,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6737434566020966,
      "step": 777
    },
    {
      "completion_length": 113.203125,
      "epoch": 0.501611863313991,
      "grad_norm": 27.09701156616211,
      "kl": 0.121337890625,
      "learning_rate": 7.491940683430044e-07,
      "loss": 0.0048,
      "reward": 1.5331305265426636,
      "reward_std": 0.1381199024617672,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5331305861473083,
      "step": 778
    },
    {
      "completion_length": 100.65625,
      "epoch": 0.5022566086395873,
      "grad_norm": 29.07394027709961,
      "kl": 0.1337890625,
      "learning_rate": 7.488716956802063e-07,
      "loss": 0.0054,
      "reward": 1.5475330352783203,
      "reward_std": 0.26855621486902237,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.5944079458713531,
      "step": 779
    },
    {
      "completion_length": 105.0625,
      "epoch": 0.5029013539651838,
      "grad_norm": 9.877948760986328,
      "kl": 0.115966796875,
      "learning_rate": 7.485493230174081e-07,
      "loss": 0.0046,
      "reward": 1.673143744468689,
      "reward_std": 0.08836501836776733,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.673143744468689,
      "step": 780
    },
    {
      "completion_length": 111.3125,
      "epoch": 0.5035460992907801,
      "grad_norm": 22.850160598754883,
      "kl": 0.132568359375,
      "learning_rate": 7.482269503546099e-07,
      "loss": 0.0053,
      "reward": 1.674381136894226,
      "reward_std": 0.12078357115387917,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6743811964988708,
      "step": 781
    },
    {
      "completion_length": 109.046875,
      "epoch": 0.5041908446163765,
      "grad_norm": 31.050485610961914,
      "kl": 0.158203125,
      "learning_rate": 7.479045776918117e-07,
      "loss": 0.0063,
      "reward": 1.4201094508171082,
      "reward_std": 0.1670779399573803,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.43573447316884995,
      "step": 782
    },
    {
      "completion_length": 118.078125,
      "epoch": 0.504835589941973,
      "grad_norm": 16.994108200073242,
      "kl": 0.103759765625,
      "learning_rate": 7.475822050290135e-07,
      "loss": 0.0042,
      "reward": 1.5727297067642212,
      "reward_std": 0.13664092868566513,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5727297067642212,
      "step": 783
    },
    {
      "completion_length": 132.359375,
      "epoch": 0.5054803352675693,
      "grad_norm": 14.750675201416016,
      "kl": 0.12158203125,
      "learning_rate": 7.472598323662152e-07,
      "loss": 0.0049,
      "reward": 1.4640502333641052,
      "reward_std": 0.19203101471066475,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.47967518121004105,
      "step": 784
    },
    {
      "completion_length": 111.03125,
      "epoch": 0.5061250805931657,
      "grad_norm": 12.755722045898438,
      "kl": 0.123779296875,
      "learning_rate": 7.469374597034172e-07,
      "loss": 0.005,
      "reward": 1.59541916847229,
      "reward_std": 0.08936497010290623,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5954191088676453,
      "step": 785
    },
    {
      "completion_length": 114.09375,
      "epoch": 0.5067698259187621,
      "grad_norm": 10.042961120605469,
      "kl": 0.129150390625,
      "learning_rate": 7.466150870406189e-07,
      "loss": 0.0052,
      "reward": 1.5148058533668518,
      "reward_std": 0.11729194968938828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5148058235645294,
      "step": 786
    },
    {
      "completion_length": 113.40625,
      "epoch": 0.5074145712443585,
      "grad_norm": 20.501480102539062,
      "kl": 0.107177734375,
      "learning_rate": 7.462927143778207e-07,
      "loss": 0.0043,
      "reward": 1.5762805938720703,
      "reward_std": 0.13149290531873703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5762805938720703,
      "step": 787
    },
    {
      "completion_length": 117.046875,
      "epoch": 0.5080593165699548,
      "grad_norm": 16.187808990478516,
      "kl": 0.100341796875,
      "learning_rate": 7.459703417150225e-07,
      "loss": 0.004,
      "reward": 1.6261343955993652,
      "reward_std": 0.13067299500107765,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6261343359947205,
      "step": 788
    },
    {
      "completion_length": 111.359375,
      "epoch": 0.5087040618955513,
      "grad_norm": 14.373567581176758,
      "kl": 0.124267578125,
      "learning_rate": 7.456479690522244e-07,
      "loss": 0.005,
      "reward": 1.4204865097999573,
      "reward_std": 0.1222645752131939,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42048661410808563,
      "step": 789
    },
    {
      "completion_length": 121.5,
      "epoch": 0.5093488072211476,
      "grad_norm": 10.425048828125,
      "kl": 0.095947265625,
      "learning_rate": 7.453255963894261e-07,
      "loss": 0.0038,
      "reward": 1.6218205690383911,
      "reward_std": 0.13687458634376526,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6218205988407135,
      "step": 790
    },
    {
      "completion_length": 125.453125,
      "epoch": 0.5099935525467441,
      "grad_norm": 11.841174125671387,
      "kl": 0.14306640625,
      "learning_rate": 7.45003223726628e-07,
      "loss": 0.0057,
      "reward": 1.548657238483429,
      "reward_std": 0.12894482910633087,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5486572682857513,
      "step": 791
    },
    {
      "completion_length": 113.015625,
      "epoch": 0.5106382978723404,
      "grad_norm": 7.590851783752441,
      "kl": 0.113525390625,
      "learning_rate": 7.446808510638297e-07,
      "loss": 0.0045,
      "reward": 1.668638288974762,
      "reward_std": 0.11230403184890747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.668638288974762,
      "step": 792
    },
    {
      "completion_length": 122.578125,
      "epoch": 0.5112830431979368,
      "grad_norm": 9.391864776611328,
      "kl": 0.1201171875,
      "learning_rate": 7.443584784010316e-07,
      "loss": 0.0048,
      "reward": 1.6186297535896301,
      "reward_std": 0.09934529662132263,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6186297535896301,
      "step": 793
    },
    {
      "completion_length": 113.65625,
      "epoch": 0.5119277885235332,
      "grad_norm": 14.11962890625,
      "kl": 0.112060546875,
      "learning_rate": 7.440361057382334e-07,
      "loss": 0.0045,
      "reward": 1.5052703619003296,
      "reward_std": 0.09183266013860703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5052704364061356,
      "step": 794
    },
    {
      "completion_length": 119.515625,
      "epoch": 0.5125725338491296,
      "grad_norm": 10.853915214538574,
      "kl": 0.13818359375,
      "learning_rate": 7.437137330754352e-07,
      "loss": 0.0055,
      "reward": 1.680991291999817,
      "reward_std": 0.14312323927879333,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6809913218021393,
      "step": 795
    },
    {
      "completion_length": 119.28125,
      "epoch": 0.513217279174726,
      "grad_norm": 42.03529739379883,
      "kl": 0.13427734375,
      "learning_rate": 7.433913604126369e-07,
      "loss": 0.0054,
      "reward": 1.6375518441200256,
      "reward_std": 0.13579606264829636,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6375519037246704,
      "step": 796
    },
    {
      "completion_length": 126.59375,
      "epoch": 0.5138620245003224,
      "grad_norm": 9.108304977416992,
      "kl": 0.12060546875,
      "learning_rate": 7.430689877498387e-07,
      "loss": 0.0048,
      "reward": 1.638499677181244,
      "reward_std": 0.09739596769213676,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6384996771812439,
      "step": 797
    },
    {
      "completion_length": 122.890625,
      "epoch": 0.5145067698259188,
      "grad_norm": 10.199109077453613,
      "kl": 0.112060546875,
      "learning_rate": 7.427466150870406e-07,
      "loss": 0.0045,
      "reward": 1.5257828831672668,
      "reward_std": 0.17206069082021713,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5414078831672668,
      "step": 798
    },
    {
      "completion_length": 112.0,
      "epoch": 0.5151515151515151,
      "grad_norm": 12.440640449523926,
      "kl": 0.16015625,
      "learning_rate": 7.424242424242424e-07,
      "loss": 0.0064,
      "reward": 1.5461509823799133,
      "reward_std": 0.1321653015911579,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5617759674787521,
      "step": 799
    },
    {
      "completion_length": 117.46875,
      "epoch": 0.5157962604771116,
      "grad_norm": 10.767664909362793,
      "kl": 0.1298828125,
      "learning_rate": 7.421018697614441e-07,
      "loss": 0.0052,
      "reward": 1.5796707272529602,
      "reward_std": 0.1553436890244484,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5952958464622498,
      "step": 800
    },
    {
      "completion_length": 124.828125,
      "epoch": 0.5164410058027079,
      "grad_norm": 12.43638801574707,
      "kl": 0.138671875,
      "learning_rate": 7.41779497098646e-07,
      "loss": 0.0055,
      "reward": 1.5580316185951233,
      "reward_std": 0.1056671105325222,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5580316483974457,
      "step": 801
    },
    {
      "completion_length": 120.140625,
      "epoch": 0.5170857511283043,
      "grad_norm": 15.19301986694336,
      "kl": 0.119384765625,
      "learning_rate": 7.414571244358478e-07,
      "loss": 0.0048,
      "reward": 1.623548150062561,
      "reward_std": 0.17357219755649567,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.623548150062561,
      "step": 802
    },
    {
      "completion_length": 123.734375,
      "epoch": 0.5177304964539007,
      "grad_norm": 18.191144943237305,
      "kl": 0.10546875,
      "learning_rate": 7.411347517730496e-07,
      "loss": 0.0042,
      "reward": 1.7189044952392578,
      "reward_std": 0.10732979327440262,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7189044952392578,
      "step": 803
    },
    {
      "completion_length": 125.03125,
      "epoch": 0.5183752417794971,
      "grad_norm": 11.427577018737793,
      "kl": 0.103515625,
      "learning_rate": 7.408123791102514e-07,
      "loss": 0.0041,
      "reward": 1.6487030982971191,
      "reward_std": 0.08015275374054909,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6487032175064087,
      "step": 804
    },
    {
      "completion_length": 111.125,
      "epoch": 0.5190199871050934,
      "grad_norm": 9.316217422485352,
      "kl": 0.154296875,
      "learning_rate": 7.404900064474532e-07,
      "loss": 0.0062,
      "reward": 1.3404886722564697,
      "reward_std": 0.13357722014188766,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.35611365735530853,
      "step": 805
    },
    {
      "completion_length": 105.34375,
      "epoch": 0.5196647324306899,
      "grad_norm": 8.146766662597656,
      "kl": 0.130126953125,
      "learning_rate": 7.40167633784655e-07,
      "loss": 0.0052,
      "reward": 1.577560842037201,
      "reward_std": 0.08118342608213425,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5775608718395233,
      "step": 806
    },
    {
      "completion_length": 108.375,
      "epoch": 0.5203094777562862,
      "grad_norm": 13.145628929138184,
      "kl": 0.11572265625,
      "learning_rate": 7.398452611218569e-07,
      "loss": 0.0046,
      "reward": 1.569650411605835,
      "reward_std": 0.11123111471533775,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5696503669023514,
      "step": 807
    },
    {
      "completion_length": 123.515625,
      "epoch": 0.5209542230818827,
      "grad_norm": 17.439918518066406,
      "kl": 0.106201171875,
      "learning_rate": 7.395228884590586e-07,
      "loss": 0.0043,
      "reward": 1.6000871062278748,
      "reward_std": 0.16571377217769623,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6157121360301971,
      "step": 808
    },
    {
      "completion_length": 123.78125,
      "epoch": 0.521598968407479,
      "grad_norm": 23.56669044494629,
      "kl": 0.126953125,
      "learning_rate": 7.392005157962604e-07,
      "loss": 0.0051,
      "reward": 1.7089502811431885,
      "reward_std": 0.09575696289539337,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7089502811431885,
      "step": 809
    },
    {
      "completion_length": 123.140625,
      "epoch": 0.5222437137330754,
      "grad_norm": 35.4520149230957,
      "kl": 0.128662109375,
      "learning_rate": 7.388781431334622e-07,
      "loss": 0.0051,
      "reward": 1.6607640981674194,
      "reward_std": 0.19404253363609314,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6607640981674194,
      "step": 810
    },
    {
      "completion_length": 106.59375,
      "epoch": 0.5228884590586719,
      "grad_norm": 10.3298978805542,
      "kl": 0.096923828125,
      "learning_rate": 7.385557704706641e-07,
      "loss": 0.0039,
      "reward": 1.455498218536377,
      "reward_std": 0.1199144497513771,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45549826323986053,
      "step": 811
    },
    {
      "completion_length": 123.015625,
      "epoch": 0.5235332043842682,
      "grad_norm": 16.133468627929688,
      "kl": 0.1884765625,
      "learning_rate": 7.382333978078658e-07,
      "loss": 0.0075,
      "reward": 1.6502628922462463,
      "reward_std": 0.13473164662718773,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.650262862443924,
      "step": 812
    },
    {
      "completion_length": 132.265625,
      "epoch": 0.5241779497098646,
      "grad_norm": 9.389092445373535,
      "kl": 0.110595703125,
      "learning_rate": 7.379110251450677e-07,
      "loss": 0.0044,
      "reward": 1.6649059057235718,
      "reward_std": 0.104056216776371,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6649059057235718,
      "step": 813
    },
    {
      "completion_length": 117.0,
      "epoch": 0.524822695035461,
      "grad_norm": 31.27338218688965,
      "kl": 0.1220703125,
      "learning_rate": 7.375886524822694e-07,
      "loss": 0.0049,
      "reward": 1.5679078698158264,
      "reward_std": 0.15254487842321396,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5679078996181488,
      "step": 814
    },
    {
      "completion_length": 134.265625,
      "epoch": 0.5254674403610574,
      "grad_norm": 13.405317306518555,
      "kl": 0.11083984375,
      "learning_rate": 7.372662798194713e-07,
      "loss": 0.0044,
      "reward": 1.5193554759025574,
      "reward_std": 0.09003760293126106,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.519355446100235,
      "step": 815
    },
    {
      "completion_length": 129.1875,
      "epoch": 0.5261121856866537,
      "grad_norm": 15.608565330505371,
      "kl": 0.113037109375,
      "learning_rate": 7.369439071566731e-07,
      "loss": 0.0045,
      "reward": 1.6685249209403992,
      "reward_std": 0.09597417712211609,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6685249507427216,
      "step": 816
    },
    {
      "completion_length": 115.84375,
      "epoch": 0.5267569310122502,
      "grad_norm": 9.022123336791992,
      "kl": 0.1123046875,
      "learning_rate": 7.366215344938749e-07,
      "loss": 0.0045,
      "reward": 1.4091280698776245,
      "reward_std": 0.0891311950981617,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4091280996799469,
      "step": 817
    },
    {
      "completion_length": 132.15625,
      "epoch": 0.5274016763378465,
      "grad_norm": 16.895809173583984,
      "kl": 0.114990234375,
      "learning_rate": 7.362991618310766e-07,
      "loss": 0.0046,
      "reward": 1.518994688987732,
      "reward_std": 0.1304328851401806,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5189946293830872,
      "step": 818
    },
    {
      "completion_length": 114.328125,
      "epoch": 0.528046421663443,
      "grad_norm": 10.180063247680664,
      "kl": 0.12353515625,
      "learning_rate": 7.359767891682786e-07,
      "loss": 0.0049,
      "reward": 1.5339825749397278,
      "reward_std": 0.14155717194080353,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5339825749397278,
      "step": 819
    },
    {
      "completion_length": 132.6875,
      "epoch": 0.5286911669890393,
      "grad_norm": 15.447187423706055,
      "kl": 0.104736328125,
      "learning_rate": 7.356544165054803e-07,
      "loss": 0.0042,
      "reward": 1.4741751551628113,
      "reward_std": 0.09367522969841957,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47417517751455307,
      "step": 820
    },
    {
      "completion_length": 123.796875,
      "epoch": 0.5293359123146357,
      "grad_norm": 7.803520202636719,
      "kl": 0.119384765625,
      "learning_rate": 7.353320438426821e-07,
      "loss": 0.0048,
      "reward": 1.5355405807495117,
      "reward_std": 0.04676287993788719,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5355406105518341,
      "step": 821
    },
    {
      "completion_length": 127.78125,
      "epoch": 0.5299806576402321,
      "grad_norm": 17.904590606689453,
      "kl": 0.1162109375,
      "learning_rate": 7.350096711798838e-07,
      "loss": 0.0046,
      "reward": 1.5604600310325623,
      "reward_std": 0.07631072774529457,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5604600012302399,
      "step": 822
    },
    {
      "completion_length": 124.8125,
      "epoch": 0.5306254029658285,
      "grad_norm": 13.199056625366211,
      "kl": 0.105712890625,
      "learning_rate": 7.346872985170857e-07,
      "loss": 0.0042,
      "reward": 1.3325340151786804,
      "reward_std": 0.055142441764473915,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3325340151786804,
      "step": 823
    },
    {
      "completion_length": 125.21875,
      "epoch": 0.5312701482914249,
      "grad_norm": 12.559577941894531,
      "kl": 0.098388671875,
      "learning_rate": 7.343649258542875e-07,
      "loss": 0.0039,
      "reward": 1.5847796201705933,
      "reward_std": 0.08421425521373749,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5847796201705933,
      "step": 824
    },
    {
      "completion_length": 120.46875,
      "epoch": 0.5319148936170213,
      "grad_norm": 9.962425231933594,
      "kl": 0.13232421875,
      "learning_rate": 7.340425531914893e-07,
      "loss": 0.0053,
      "reward": 1.7481776475906372,
      "reward_std": 0.14498769491910934,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7481776475906372,
      "step": 825
    },
    {
      "completion_length": 117.0,
      "epoch": 0.5325596389426177,
      "grad_norm": 12.92812442779541,
      "kl": 0.114013671875,
      "learning_rate": 7.337201805286911e-07,
      "loss": 0.0046,
      "reward": 1.5807170271873474,
      "reward_std": 0.14974131435155869,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5807170867919922,
      "step": 826
    },
    {
      "completion_length": 116.578125,
      "epoch": 0.533204384268214,
      "grad_norm": 15.796828269958496,
      "kl": 0.113037109375,
      "learning_rate": 7.333978078658929e-07,
      "loss": 0.0045,
      "reward": 1.6642200946807861,
      "reward_std": 0.11426962912082672,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6642200648784637,
      "step": 827
    },
    {
      "completion_length": 110.4375,
      "epoch": 0.5338491295938105,
      "grad_norm": 38.065834045410156,
      "kl": 0.12548828125,
      "learning_rate": 7.330754352030947e-07,
      "loss": 0.005,
      "reward": 1.557353138923645,
      "reward_std": 0.0908035859465599,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5573531538248062,
      "step": 828
    },
    {
      "completion_length": 121.828125,
      "epoch": 0.5344938749194068,
      "grad_norm": 11.057354927062988,
      "kl": 0.1044921875,
      "learning_rate": 7.327530625402966e-07,
      "loss": 0.0042,
      "reward": 1.5736215710639954,
      "reward_std": 0.1010180115699768,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5736216008663177,
      "step": 829
    },
    {
      "completion_length": 120.046875,
      "epoch": 0.5351386202450032,
      "grad_norm": 24.34914207458496,
      "kl": 0.1357421875,
      "learning_rate": 7.324306898774983e-07,
      "loss": 0.0054,
      "reward": 1.3760216236114502,
      "reward_std": 0.05299963988363743,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.37602153420448303,
      "step": 830
    },
    {
      "completion_length": 118.734375,
      "epoch": 0.5357833655705996,
      "grad_norm": 10.806406021118164,
      "kl": 0.1083984375,
      "learning_rate": 7.321083172147001e-07,
      "loss": 0.0043,
      "reward": 1.3271194100379944,
      "reward_std": 0.04556073807179928,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3271194323897362,
      "step": 831
    },
    {
      "completion_length": 128.125,
      "epoch": 0.536428110896196,
      "grad_norm": 16.560606002807617,
      "kl": 0.12060546875,
      "learning_rate": 7.31785944551902e-07,
      "loss": 0.0048,
      "reward": 1.6746159791946411,
      "reward_std": 0.09150423854589462,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6746159195899963,
      "step": 832
    },
    {
      "completion_length": 129.953125,
      "epoch": 0.5370728562217923,
      "grad_norm": 14.166282653808594,
      "kl": 0.1298828125,
      "learning_rate": 7.314635718891038e-07,
      "loss": 0.0052,
      "reward": 1.7240188121795654,
      "reward_std": 0.11212188005447388,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7240188419818878,
      "step": 833
    },
    {
      "completion_length": 124.140625,
      "epoch": 0.5377176015473888,
      "grad_norm": 46.64698028564453,
      "kl": 0.11962890625,
      "learning_rate": 7.311411992263055e-07,
      "loss": 0.0048,
      "reward": 1.6080630421638489,
      "reward_std": 0.11205699667334557,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6080630421638489,
      "step": 834
    },
    {
      "completion_length": 119.90625,
      "epoch": 0.5383623468729851,
      "grad_norm": 8.266460418701172,
      "kl": 0.10791015625,
      "learning_rate": 7.308188265635074e-07,
      "loss": 0.0043,
      "reward": 1.5985249280929565,
      "reward_std": 0.11095097288489342,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5985249280929565,
      "step": 835
    },
    {
      "completion_length": 121.65625,
      "epoch": 0.5390070921985816,
      "grad_norm": 47.86563491821289,
      "kl": 0.117919921875,
      "learning_rate": 7.304964539007092e-07,
      "loss": 0.0047,
      "reward": 1.4885122179985046,
      "reward_std": 0.07481271028518677,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48851217329502106,
      "step": 836
    },
    {
      "completion_length": 125.828125,
      "epoch": 0.539651837524178,
      "grad_norm": 21.55581283569336,
      "kl": 0.12255859375,
      "learning_rate": 7.30174081237911e-07,
      "loss": 0.0049,
      "reward": 1.594322383403778,
      "reward_std": 0.14254087209701538,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5943224132061005,
      "step": 837
    },
    {
      "completion_length": 114.328125,
      "epoch": 0.5402965828497743,
      "grad_norm": 16.470298767089844,
      "kl": 0.14453125,
      "learning_rate": 7.298517085751128e-07,
      "loss": 0.0058,
      "reward": 1.6632260084152222,
      "reward_std": 0.12153082340955734,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6632260978221893,
      "step": 838
    },
    {
      "completion_length": 116.84375,
      "epoch": 0.5409413281753708,
      "grad_norm": 10.931767463684082,
      "kl": 0.108154296875,
      "learning_rate": 7.295293359123146e-07,
      "loss": 0.0043,
      "reward": 1.6424379348754883,
      "reward_std": 0.15779002383351326,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6580629050731659,
      "step": 839
    },
    {
      "completion_length": 111.703125,
      "epoch": 0.5415860735009671,
      "grad_norm": 13.211544036865234,
      "kl": 0.112060546875,
      "learning_rate": 7.292069632495163e-07,
      "loss": 0.0045,
      "reward": 1.555619239807129,
      "reward_std": 0.11191636323928833,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5712441802024841,
      "step": 840
    },
    {
      "completion_length": 110.90625,
      "epoch": 0.5422308188265635,
      "grad_norm": 9.324379920959473,
      "kl": 0.145263671875,
      "learning_rate": 7.288845905867183e-07,
      "loss": 0.0058,
      "reward": 1.5834694504737854,
      "reward_std": 0.09625018388032913,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5834693908691406,
      "step": 841
    },
    {
      "completion_length": 121.515625,
      "epoch": 0.5428755641521599,
      "grad_norm": 17.784393310546875,
      "kl": 0.15380859375,
      "learning_rate": 7.2856221792392e-07,
      "loss": 0.0062,
      "reward": 1.4552746415138245,
      "reward_std": 0.10180060565471649,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45527467131614685,
      "step": 842
    },
    {
      "completion_length": 113.875,
      "epoch": 0.5435203094777563,
      "grad_norm": 14.1525239944458,
      "kl": 0.1171875,
      "learning_rate": 7.282398452611218e-07,
      "loss": 0.0047,
      "reward": 1.5815837383270264,
      "reward_std": 0.06838434562087059,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5815837979316711,
      "step": 843
    },
    {
      "completion_length": 111.8125,
      "epoch": 0.5441650548033526,
      "grad_norm": 14.385869026184082,
      "kl": 0.134765625,
      "learning_rate": 7.279174725983235e-07,
      "loss": 0.0054,
      "reward": 1.5503296852111816,
      "reward_std": 0.08688557147979736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5503296554088593,
      "step": 844
    },
    {
      "completion_length": 108.9375,
      "epoch": 0.5448098001289491,
      "grad_norm": 13.819561958312988,
      "kl": 0.1298828125,
      "learning_rate": 7.275950999355255e-07,
      "loss": 0.0052,
      "reward": 1.3759878873825073,
      "reward_std": 0.07583236694335938,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3759878873825073,
      "step": 845
    },
    {
      "completion_length": 113.078125,
      "epoch": 0.5454545454545454,
      "grad_norm": 13.309516906738281,
      "kl": 0.11767578125,
      "learning_rate": 7.272727272727272e-07,
      "loss": 0.0047,
      "reward": 1.5513026714324951,
      "reward_std": 0.0639763344079256,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5513026863336563,
      "step": 846
    },
    {
      "completion_length": 125.8125,
      "epoch": 0.5460992907801419,
      "grad_norm": 21.36383056640625,
      "kl": 0.115966796875,
      "learning_rate": 7.26950354609929e-07,
      "loss": 0.0046,
      "reward": 1.4737462997436523,
      "reward_std": 0.11164987832307816,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47374628484249115,
      "step": 847
    },
    {
      "completion_length": 132.8125,
      "epoch": 0.5467440361057382,
      "grad_norm": 10.429849624633789,
      "kl": 0.107177734375,
      "learning_rate": 7.266279819471308e-07,
      "loss": 0.0043,
      "reward": 1.6067231893539429,
      "reward_std": 0.10046634450554848,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6067231893539429,
      "step": 848
    },
    {
      "completion_length": 118.578125,
      "epoch": 0.5473887814313346,
      "grad_norm": 20.214109420776367,
      "kl": 0.13818359375,
      "learning_rate": 7.263056092843327e-07,
      "loss": 0.0055,
      "reward": 1.4823689460754395,
      "reward_std": 0.10195580497384071,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48236894607543945,
      "step": 849
    },
    {
      "completion_length": 105.421875,
      "epoch": 0.5480335267569311,
      "grad_norm": 15.529704093933105,
      "kl": 0.146484375,
      "learning_rate": 7.259832366215344e-07,
      "loss": 0.0059,
      "reward": 1.6683238744735718,
      "reward_std": 0.13429173827171326,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.668323814868927,
      "step": 850
    },
    {
      "completion_length": 107.359375,
      "epoch": 0.5486782720825274,
      "grad_norm": 15.431048393249512,
      "kl": 0.109375,
      "learning_rate": 7.256608639587363e-07,
      "loss": 0.0044,
      "reward": 1.4428613781929016,
      "reward_std": 0.046914756298065186,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44286131858825684,
      "step": 851
    },
    {
      "completion_length": 104.890625,
      "epoch": 0.5493230174081238,
      "grad_norm": 20.874425888061523,
      "kl": 0.14306640625,
      "learning_rate": 7.25338491295938e-07,
      "loss": 0.0057,
      "reward": 1.5059730410575867,
      "reward_std": 0.10780579224228859,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5059730112552643,
      "step": 852
    },
    {
      "completion_length": 122.0625,
      "epoch": 0.5499677627337202,
      "grad_norm": 9.9263277053833,
      "kl": 0.122314453125,
      "learning_rate": 7.250161186331398e-07,
      "loss": 0.0049,
      "reward": 1.385474145412445,
      "reward_std": 0.06568328104913235,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3854740560054779,
      "step": 853
    },
    {
      "completion_length": 119.046875,
      "epoch": 0.5506125080593166,
      "grad_norm": 33.775062561035156,
      "kl": 0.152587890625,
      "learning_rate": 7.246937459703417e-07,
      "loss": 0.0061,
      "reward": 1.6085370779037476,
      "reward_std": 0.14623812586069107,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6085370779037476,
      "step": 854
    },
    {
      "completion_length": 126.625,
      "epoch": 0.5512572533849129,
      "grad_norm": 13.095032691955566,
      "kl": 0.11376953125,
      "learning_rate": 7.243713733075435e-07,
      "loss": 0.0046,
      "reward": 1.6513434052467346,
      "reward_std": 0.14508388936519623,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.666968435049057,
      "step": 855
    },
    {
      "completion_length": 113.0,
      "epoch": 0.5519019987105094,
      "grad_norm": 14.717634201049805,
      "kl": 0.113037109375,
      "learning_rate": 7.240490006447452e-07,
      "loss": 0.0045,
      "reward": 1.5023114681243896,
      "reward_std": 0.07279381342232227,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5023114830255508,
      "step": 856
    },
    {
      "completion_length": 125.484375,
      "epoch": 0.5525467440361057,
      "grad_norm": 12.679544448852539,
      "kl": 0.113525390625,
      "learning_rate": 7.237266279819471e-07,
      "loss": 0.0045,
      "reward": 1.5734071135520935,
      "reward_std": 0.052911147475242615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5734071433544159,
      "step": 857
    },
    {
      "completion_length": 112.21875,
      "epoch": 0.5531914893617021,
      "grad_norm": 22.410959243774414,
      "kl": 0.10791015625,
      "learning_rate": 7.23404255319149e-07,
      "loss": 0.0043,
      "reward": 1.518027424812317,
      "reward_std": 0.054683832451701164,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5180274546146393,
      "step": 858
    },
    {
      "completion_length": 117.359375,
      "epoch": 0.5538362346872985,
      "grad_norm": 8.975079536437988,
      "kl": 0.11083984375,
      "learning_rate": 7.230818826563507e-07,
      "loss": 0.0044,
      "reward": 1.531017243862152,
      "reward_std": 0.1474187821149826,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5466422140598297,
      "step": 859
    },
    {
      "completion_length": 120.359375,
      "epoch": 0.5544809800128949,
      "grad_norm": 30.06081199645996,
      "kl": 0.125,
      "learning_rate": 7.227595099935525e-07,
      "loss": 0.005,
      "reward": 1.472097635269165,
      "reward_std": 0.12772921100258827,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4720976650714874,
      "step": 860
    },
    {
      "completion_length": 121.296875,
      "epoch": 0.5551257253384912,
      "grad_norm": 7.828312397003174,
      "kl": 0.111328125,
      "learning_rate": 7.224371373307543e-07,
      "loss": 0.0045,
      "reward": 1.5114521980285645,
      "reward_std": 0.09453960880637169,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5114522278308868,
      "step": 861
    },
    {
      "completion_length": 120.09375,
      "epoch": 0.5557704706640877,
      "grad_norm": 42.72480773925781,
      "kl": 0.1083984375,
      "learning_rate": 7.221147646679562e-07,
      "loss": 0.0043,
      "reward": 1.5521915555000305,
      "reward_std": 0.07748100534081459,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5521914958953857,
      "step": 862
    },
    {
      "completion_length": 127.5625,
      "epoch": 0.556415215989684,
      "grad_norm": 14.040475845336914,
      "kl": 0.143310546875,
      "learning_rate": 7.21792392005158e-07,
      "loss": 0.0057,
      "reward": 1.5281696319580078,
      "reward_std": 0.10375509038567543,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.528169646859169,
      "step": 863
    },
    {
      "completion_length": 125.4375,
      "epoch": 0.5570599613152805,
      "grad_norm": 10.69442367553711,
      "kl": 0.123046875,
      "learning_rate": 7.214700193423598e-07,
      "loss": 0.0049,
      "reward": 1.612160563468933,
      "reward_std": 0.14196538925170898,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6277855336666107,
      "step": 864
    },
    {
      "completion_length": 113.03125,
      "epoch": 0.5577047066408769,
      "grad_norm": 12.548157691955566,
      "kl": 0.093017578125,
      "learning_rate": 7.211476466795615e-07,
      "loss": 0.0037,
      "reward": 1.5996488332748413,
      "reward_std": 0.08111307956278324,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5996488332748413,
      "step": 865
    },
    {
      "completion_length": 118.4375,
      "epoch": 0.5583494519664732,
      "grad_norm": 19.041934967041016,
      "kl": 0.10791015625,
      "learning_rate": 7.208252740167632e-07,
      "loss": 0.0043,
      "reward": 1.5888526439666748,
      "reward_std": 0.10432974994182587,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5888526439666748,
      "step": 866
    },
    {
      "completion_length": 104.59375,
      "epoch": 0.5589941972920697,
      "grad_norm": 44.237117767333984,
      "kl": 0.1376953125,
      "learning_rate": 7.205029013539652e-07,
      "loss": 0.0055,
      "reward": 1.6344898343086243,
      "reward_std": 0.08369472622871399,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6344898343086243,
      "step": 867
    },
    {
      "completion_length": 100.75,
      "epoch": 0.559638942617666,
      "grad_norm": 12.805983543395996,
      "kl": 0.146240234375,
      "learning_rate": 7.20180528691167e-07,
      "loss": 0.0058,
      "reward": 1.6388565301895142,
      "reward_std": 0.055887432768940926,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6388565599918365,
      "step": 868
    },
    {
      "completion_length": 111.90625,
      "epoch": 0.5602836879432624,
      "grad_norm": 14.531253814697266,
      "kl": 0.115478515625,
      "learning_rate": 7.198581560283687e-07,
      "loss": 0.0046,
      "reward": 1.6537020206451416,
      "reward_std": 0.10696634277701378,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.653702050447464,
      "step": 869
    },
    {
      "completion_length": 107.546875,
      "epoch": 0.5609284332688588,
      "grad_norm": 18.164827346801758,
      "kl": 0.13525390625,
      "learning_rate": 7.195357833655706e-07,
      "loss": 0.0054,
      "reward": 1.6144909858703613,
      "reward_std": 0.1230674758553505,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6144909858703613,
      "step": 870
    },
    {
      "completion_length": 103.453125,
      "epoch": 0.5615731785944552,
      "grad_norm": 14.088841438293457,
      "kl": 0.13232421875,
      "learning_rate": 7.192134107027724e-07,
      "loss": 0.0053,
      "reward": 1.6711689233779907,
      "reward_std": 0.10582815483212471,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.671168863773346,
      "step": 871
    },
    {
      "completion_length": 100.109375,
      "epoch": 0.5622179239200515,
      "grad_norm": 17.895931243896484,
      "kl": 0.11865234375,
      "learning_rate": 7.188910380399742e-07,
      "loss": 0.0048,
      "reward": 1.6541008949279785,
      "reward_std": 0.14635222405195236,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6697258949279785,
      "step": 872
    },
    {
      "completion_length": 101.234375,
      "epoch": 0.562862669245648,
      "grad_norm": 11.591320991516113,
      "kl": 0.138916015625,
      "learning_rate": 7.18568665377176e-07,
      "loss": 0.0055,
      "reward": 1.6059136390686035,
      "reward_std": 0.08753130957484245,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6059136092662811,
      "step": 873
    },
    {
      "completion_length": 103.953125,
      "epoch": 0.5635074145712443,
      "grad_norm": 10.312195777893066,
      "kl": 0.151123046875,
      "learning_rate": 7.182462927143778e-07,
      "loss": 0.006,
      "reward": 1.5342779755592346,
      "reward_std": 0.05523178353905678,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5342779606580734,
      "step": 874
    },
    {
      "completion_length": 106.34375,
      "epoch": 0.5641521598968408,
      "grad_norm": 240.9861602783203,
      "kl": 0.134521484375,
      "learning_rate": 7.179239200515796e-07,
      "loss": 0.0054,
      "reward": 1.5363209247589111,
      "reward_std": 0.12755867093801498,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5519458651542664,
      "step": 875
    },
    {
      "completion_length": 99.359375,
      "epoch": 0.5647969052224371,
      "grad_norm": 26.704421997070312,
      "kl": 0.14892578125,
      "learning_rate": 7.176015473887815e-07,
      "loss": 0.006,
      "reward": 1.6137511134147644,
      "reward_std": 0.0752146728336811,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6137511432170868,
      "step": 876
    },
    {
      "completion_length": 117.0,
      "epoch": 0.5654416505480335,
      "grad_norm": 15.467166900634766,
      "kl": 0.22119140625,
      "learning_rate": 7.172791747259832e-07,
      "loss": 0.0088,
      "reward": 1.5769726037979126,
      "reward_std": 0.12701868265867233,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5769726037979126,
      "step": 877
    },
    {
      "completion_length": 108.671875,
      "epoch": 0.56608639587363,
      "grad_norm": 18.265071868896484,
      "kl": 0.115478515625,
      "learning_rate": 7.16956802063185e-07,
      "loss": 0.0046,
      "reward": 1.6923354864120483,
      "reward_std": 0.07585858926177025,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6923355162143707,
      "step": 878
    },
    {
      "completion_length": 113.578125,
      "epoch": 0.5667311411992263,
      "grad_norm": 9.981937408447266,
      "kl": 0.120361328125,
      "learning_rate": 7.166344294003868e-07,
      "loss": 0.0048,
      "reward": 1.5243776440620422,
      "reward_std": 0.09604435786604881,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.524377629160881,
      "step": 879
    },
    {
      "completion_length": 114.0625,
      "epoch": 0.5673758865248227,
      "grad_norm": 24.6324520111084,
      "kl": 0.20849609375,
      "learning_rate": 7.163120567375887e-07,
      "loss": 0.0084,
      "reward": 1.6049633622169495,
      "reward_std": 0.08231936953961849,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6049633026123047,
      "step": 880
    },
    {
      "completion_length": 109.015625,
      "epoch": 0.5680206318504191,
      "grad_norm": 10.54612922668457,
      "kl": 0.1181640625,
      "learning_rate": 7.159896840747904e-07,
      "loss": 0.0047,
      "reward": 1.2566799521446228,
      "reward_std": 0.06597771868109703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2566799595952034,
      "step": 881
    },
    {
      "completion_length": 113.859375,
      "epoch": 0.5686653771760155,
      "grad_norm": 11.331385612487793,
      "kl": 0.14111328125,
      "learning_rate": 7.156673114119923e-07,
      "loss": 0.0056,
      "reward": 1.6598809957504272,
      "reward_std": 0.08520251885056496,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6598810255527496,
      "step": 882
    },
    {
      "completion_length": 115.703125,
      "epoch": 0.5693101225016118,
      "grad_norm": 16.860124588012695,
      "kl": 0.10009765625,
      "learning_rate": 7.15344938749194e-07,
      "loss": 0.004,
      "reward": 1.6775421500205994,
      "reward_std": 0.09860784560441971,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6775421500205994,
      "step": 883
    },
    {
      "completion_length": 103.21875,
      "epoch": 0.5699548678272083,
      "grad_norm": 33.44534683227539,
      "kl": 0.13330078125,
      "learning_rate": 7.150225660863959e-07,
      "loss": 0.0053,
      "reward": 1.5252437591552734,
      "reward_std": 0.06859505549073219,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5252437740564346,
      "step": 884
    },
    {
      "completion_length": 117.265625,
      "epoch": 0.5705996131528046,
      "grad_norm": 12.154504776000977,
      "kl": 0.137451171875,
      "learning_rate": 7.147001934235977e-07,
      "loss": 0.0055,
      "reward": 1.6008020639419556,
      "reward_std": 0.12660085037350655,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6008020639419556,
      "step": 885
    },
    {
      "completion_length": 121.328125,
      "epoch": 0.571244358478401,
      "grad_norm": 18.56639862060547,
      "kl": 0.11767578125,
      "learning_rate": 7.143778207607995e-07,
      "loss": 0.0047,
      "reward": 1.6193873286247253,
      "reward_std": 0.10898285359144211,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.619387298822403,
      "step": 886
    },
    {
      "completion_length": 130.421875,
      "epoch": 0.5718891038039974,
      "grad_norm": 174.48056030273438,
      "kl": 0.123291015625,
      "learning_rate": 7.140554480980012e-07,
      "loss": 0.0049,
      "reward": 1.5258971452713013,
      "reward_std": 0.05683533661067486,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5258970856666565,
      "step": 887
    },
    {
      "completion_length": 111.125,
      "epoch": 0.5725338491295938,
      "grad_norm": 7.736058712005615,
      "kl": 0.114501953125,
      "learning_rate": 7.137330754352032e-07,
      "loss": 0.0046,
      "reward": 1.6826195120811462,
      "reward_std": 0.13724345713853836,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6982445120811462,
      "step": 888
    },
    {
      "completion_length": 120.171875,
      "epoch": 0.5731785944551901,
      "grad_norm": 45.026058197021484,
      "kl": 0.13623046875,
      "learning_rate": 7.134107027724049e-07,
      "loss": 0.0055,
      "reward": 1.5901039242744446,
      "reward_std": 0.16770201176404953,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5901040136814117,
      "step": 889
    },
    {
      "completion_length": 113.640625,
      "epoch": 0.5738233397807866,
      "grad_norm": 9.868265151977539,
      "kl": 0.126708984375,
      "learning_rate": 7.130883301096067e-07,
      "loss": 0.0051,
      "reward": 1.4987885355949402,
      "reward_std": 0.14777783304452896,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4987884759902954,
      "step": 890
    },
    {
      "completion_length": 103.046875,
      "epoch": 0.574468085106383,
      "grad_norm": 16.197307586669922,
      "kl": 0.12451171875,
      "learning_rate": 7.127659574468084e-07,
      "loss": 0.005,
      "reward": 1.5383508205413818,
      "reward_std": 0.09310048446059227,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5383508205413818,
      "step": 891
    },
    {
      "completion_length": 115.203125,
      "epoch": 0.5751128304319794,
      "grad_norm": 7.911108016967773,
      "kl": 0.133056640625,
      "learning_rate": 7.124435847840103e-07,
      "loss": 0.0053,
      "reward": 1.6043503284454346,
      "reward_std": 0.12782978266477585,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6043503284454346,
      "step": 892
    },
    {
      "completion_length": 110.71875,
      "epoch": 0.5757575757575758,
      "grad_norm": 49.204200744628906,
      "kl": 0.135498046875,
      "learning_rate": 7.121212121212121e-07,
      "loss": 0.0054,
      "reward": 1.557990849018097,
      "reward_std": 0.14290058612823486,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5579908788204193,
      "step": 893
    },
    {
      "completion_length": 119.65625,
      "epoch": 0.5764023210831721,
      "grad_norm": 66.35208129882812,
      "kl": 0.10546875,
      "learning_rate": 7.117988394584139e-07,
      "loss": 0.0042,
      "reward": 1.5070759654045105,
      "reward_std": 0.1017339937388897,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5070759654045105,
      "step": 894
    },
    {
      "completion_length": 106.046875,
      "epoch": 0.5770470664087686,
      "grad_norm": 10.324350357055664,
      "kl": 0.12451171875,
      "learning_rate": 7.114764667956157e-07,
      "loss": 0.005,
      "reward": 1.5135289430618286,
      "reward_std": 0.18148383498191833,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5291539430618286,
      "step": 895
    },
    {
      "completion_length": 121.34375,
      "epoch": 0.5776918117343649,
      "grad_norm": 25.454736709594727,
      "kl": 0.124267578125,
      "learning_rate": 7.111540941328175e-07,
      "loss": 0.005,
      "reward": 1.5388038158416748,
      "reward_std": 0.09261330217123032,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5388037711381912,
      "step": 896
    },
    {
      "completion_length": 113.0625,
      "epoch": 0.5783365570599613,
      "grad_norm": 7.527803421020508,
      "kl": 0.12060546875,
      "learning_rate": 7.108317214700193e-07,
      "loss": 0.0048,
      "reward": 1.6463514566421509,
      "reward_std": 0.1266507338732481,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6463514566421509,
      "step": 897
    },
    {
      "completion_length": 115.125,
      "epoch": 0.5789813023855577,
      "grad_norm": 51.9161262512207,
      "kl": 0.1162109375,
      "learning_rate": 7.105093488072212e-07,
      "loss": 0.0046,
      "reward": 1.5849735736846924,
      "reward_std": 0.08285956270992756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5849734991788864,
      "step": 898
    },
    {
      "completion_length": 114.75,
      "epoch": 0.5796260477111541,
      "grad_norm": 14.701598167419434,
      "kl": 0.164306640625,
      "learning_rate": 7.101869761444229e-07,
      "loss": 0.0066,
      "reward": 1.4290552139282227,
      "reward_std": 0.11476770415902138,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.44468019902706146,
      "step": 899
    },
    {
      "completion_length": 118.921875,
      "epoch": 0.5802707930367504,
      "grad_norm": 7.526551246643066,
      "kl": 0.11865234375,
      "learning_rate": 7.098646034816247e-07,
      "loss": 0.0047,
      "reward": 1.523693561553955,
      "reward_std": 0.06077193468809128,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5236935317516327,
      "step": 900
    },
    {
      "completion_length": 116.796875,
      "epoch": 0.5809155383623469,
      "grad_norm": 12.488341331481934,
      "kl": 0.1728515625,
      "learning_rate": 7.095422308188266e-07,
      "loss": 0.0069,
      "reward": 1.5757187604904175,
      "reward_std": 0.11441710963845253,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5757187008857727,
      "step": 901
    },
    {
      "completion_length": 114.453125,
      "epoch": 0.5815602836879432,
      "grad_norm": 25.193092346191406,
      "kl": 0.125,
      "learning_rate": 7.092198581560284e-07,
      "loss": 0.005,
      "reward": 1.601715087890625,
      "reward_std": 0.08856208249926567,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6017151474952698,
      "step": 902
    },
    {
      "completion_length": 124.03125,
      "epoch": 0.5822050290135397,
      "grad_norm": 11.461077690124512,
      "kl": 0.12255859375,
      "learning_rate": 7.088974854932301e-07,
      "loss": 0.0049,
      "reward": 1.4974235892295837,
      "reward_std": 0.12268773466348648,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49742357432842255,
      "step": 903
    },
    {
      "completion_length": 116.640625,
      "epoch": 0.5828497743391361,
      "grad_norm": 17.099470138549805,
      "kl": 0.130615234375,
      "learning_rate": 7.08575112830432e-07,
      "loss": 0.0052,
      "reward": 1.499584436416626,
      "reward_std": 0.05868752859532833,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4995844215154648,
      "step": 904
    },
    {
      "completion_length": 131.15625,
      "epoch": 0.5834945196647324,
      "grad_norm": 19.47797966003418,
      "kl": 0.123291015625,
      "learning_rate": 7.082527401676337e-07,
      "loss": 0.0049,
      "reward": 1.4962482452392578,
      "reward_std": 0.08062171936035156,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4962482154369354,
      "step": 905
    },
    {
      "completion_length": 114.4375,
      "epoch": 0.5841392649903289,
      "grad_norm": 12.948217391967773,
      "kl": 0.1376953125,
      "learning_rate": 7.079303675048356e-07,
      "loss": 0.0055,
      "reward": 1.5855439901351929,
      "reward_std": 0.07610942982137203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5855439305305481,
      "step": 906
    },
    {
      "completion_length": 122.375,
      "epoch": 0.5847840103159252,
      "grad_norm": 11.65310287475586,
      "kl": 0.13427734375,
      "learning_rate": 7.076079948420374e-07,
      "loss": 0.0054,
      "reward": 1.4916418194770813,
      "reward_std": 0.06596004217863083,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4916418194770813,
      "step": 907
    },
    {
      "completion_length": 122.03125,
      "epoch": 0.5854287556415216,
      "grad_norm": 9.089468955993652,
      "kl": 0.13623046875,
      "learning_rate": 7.072856221792392e-07,
      "loss": 0.0054,
      "reward": 1.5869945883750916,
      "reward_std": 0.09673259779810905,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5869945287704468,
      "step": 908
    },
    {
      "completion_length": 125.390625,
      "epoch": 0.586073500967118,
      "grad_norm": 13.814423561096191,
      "kl": 0.125732421875,
      "learning_rate": 7.069632495164409e-07,
      "loss": 0.005,
      "reward": 1.4843732714653015,
      "reward_std": 0.10544757917523384,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4843732565641403,
      "step": 909
    },
    {
      "completion_length": 118.09375,
      "epoch": 0.5867182462927144,
      "grad_norm": 12.298931121826172,
      "kl": 0.122802734375,
      "learning_rate": 7.066408768536429e-07,
      "loss": 0.0049,
      "reward": 1.576155424118042,
      "reward_std": 0.10403759777545929,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5761553943157196,
      "step": 910
    },
    {
      "completion_length": 136.484375,
      "epoch": 0.5873629916183107,
      "grad_norm": 17.558902740478516,
      "kl": 0.1494140625,
      "learning_rate": 7.063185041908446e-07,
      "loss": 0.006,
      "reward": 1.5917853116989136,
      "reward_std": 0.1398775726556778,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5917853713035583,
      "step": 911
    },
    {
      "completion_length": 117.828125,
      "epoch": 0.5880077369439072,
      "grad_norm": 14.441729545593262,
      "kl": 0.116455078125,
      "learning_rate": 7.059961315280464e-07,
      "loss": 0.0047,
      "reward": 1.538801908493042,
      "reward_std": 0.08991247788071632,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.538801908493042,
      "step": 912
    },
    {
      "completion_length": 114.21875,
      "epoch": 0.5886524822695035,
      "grad_norm": 8.804062843322754,
      "kl": 0.1318359375,
      "learning_rate": 7.056737588652481e-07,
      "loss": 0.0053,
      "reward": 1.6029040813446045,
      "reward_std": 0.1059550866484642,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6029041111469269,
      "step": 913
    },
    {
      "completion_length": 133.09375,
      "epoch": 0.5892972275951,
      "grad_norm": 31.393301010131836,
      "kl": 0.14501953125,
      "learning_rate": 7.053513862024501e-07,
      "loss": 0.0058,
      "reward": 1.5376715660095215,
      "reward_std": 0.07115558534860611,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5376715958118439,
      "step": 914
    },
    {
      "completion_length": 120.140625,
      "epoch": 0.5899419729206963,
      "grad_norm": 11.773250579833984,
      "kl": 0.115234375,
      "learning_rate": 7.050290135396518e-07,
      "loss": 0.0046,
      "reward": 1.6650592684745789,
      "reward_std": 0.061884213238954544,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6650592386722565,
      "step": 915
    },
    {
      "completion_length": 116.359375,
      "epoch": 0.5905867182462927,
      "grad_norm": 11.13023567199707,
      "kl": 0.1357421875,
      "learning_rate": 7.047066408768536e-07,
      "loss": 0.0054,
      "reward": 1.3285646438598633,
      "reward_std": 0.15595447272062302,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.3441896289587021,
      "step": 916
    },
    {
      "completion_length": 109.796875,
      "epoch": 0.591231463571889,
      "grad_norm": 17.604372024536133,
      "kl": 0.123779296875,
      "learning_rate": 7.043842682140554e-07,
      "loss": 0.005,
      "reward": 1.4047325253486633,
      "reward_std": 0.07474876940250397,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.40473249554634094,
      "step": 917
    },
    {
      "completion_length": 107.1875,
      "epoch": 0.5918762088974855,
      "grad_norm": 17.5688533782959,
      "kl": 0.130859375,
      "learning_rate": 7.040618955512572e-07,
      "loss": 0.0052,
      "reward": 1.6198034286499023,
      "reward_std": 0.15777599066495895,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6198033392429352,
      "step": 918
    },
    {
      "completion_length": 102.328125,
      "epoch": 0.5925209542230819,
      "grad_norm": 17.94205665588379,
      "kl": 0.166259765625,
      "learning_rate": 7.03739522888459e-07,
      "loss": 0.0067,
      "reward": 1.4272202253341675,
      "reward_std": 0.05302144214510918,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42722028493881226,
      "step": 919
    },
    {
      "completion_length": 104.921875,
      "epoch": 0.5931656995486783,
      "grad_norm": 14.72970962524414,
      "kl": 0.1484375,
      "learning_rate": 7.034171502256609e-07,
      "loss": 0.0059,
      "reward": 1.7611448764801025,
      "reward_std": 0.08485203981399536,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7611448764801025,
      "step": 920
    },
    {
      "completion_length": 111.671875,
      "epoch": 0.5938104448742747,
      "grad_norm": 22.33138084411621,
      "kl": 0.14404296875,
      "learning_rate": 7.030947775628626e-07,
      "loss": 0.0058,
      "reward": 1.6409417390823364,
      "reward_std": 0.10936198756098747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6409417390823364,
      "step": 921
    },
    {
      "completion_length": 106.65625,
      "epoch": 0.594455190199871,
      "grad_norm": 14.650537490844727,
      "kl": 0.134765625,
      "learning_rate": 7.027724049000644e-07,
      "loss": 0.0054,
      "reward": 1.5477882027626038,
      "reward_std": 0.10123782977461815,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5477882623672485,
      "step": 922
    },
    {
      "completion_length": 113.265625,
      "epoch": 0.5950999355254675,
      "grad_norm": 13.721552848815918,
      "kl": 0.13720703125,
      "learning_rate": 7.024500322372663e-07,
      "loss": 0.0055,
      "reward": 1.5293274521827698,
      "reward_std": 0.09637310355901718,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5293275117874146,
      "step": 923
    },
    {
      "completion_length": 104.375,
      "epoch": 0.5957446808510638,
      "grad_norm": 14.026667594909668,
      "kl": 0.116455078125,
      "learning_rate": 7.021276595744681e-07,
      "loss": 0.0047,
      "reward": 1.4343553185462952,
      "reward_std": 0.057373858988285065,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43435533344745636,
      "step": 924
    },
    {
      "completion_length": 104.265625,
      "epoch": 0.5963894261766602,
      "grad_norm": 16.487895965576172,
      "kl": 0.1181640625,
      "learning_rate": 7.018052869116698e-07,
      "loss": 0.0047,
      "reward": 1.6196611523628235,
      "reward_std": 0.08072704821825027,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6196612119674683,
      "step": 925
    },
    {
      "completion_length": 102.171875,
      "epoch": 0.5970341715022566,
      "grad_norm": 23.069528579711914,
      "kl": 0.12158203125,
      "learning_rate": 7.014829142488717e-07,
      "loss": 0.0049,
      "reward": 1.632835030555725,
      "reward_std": 0.08614074625074863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6328349560499191,
      "step": 926
    },
    {
      "completion_length": 113.53125,
      "epoch": 0.597678916827853,
      "grad_norm": 27.1119327545166,
      "kl": 0.12548828125,
      "learning_rate": 7.011605415860735e-07,
      "loss": 0.005,
      "reward": 1.4445176720619202,
      "reward_std": 0.14256715029478073,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44451767206192017,
      "step": 927
    },
    {
      "completion_length": 97.171875,
      "epoch": 0.5983236621534493,
      "grad_norm": 19.736724853515625,
      "kl": 0.13134765625,
      "learning_rate": 7.008381689232753e-07,
      "loss": 0.0052,
      "reward": 1.601191520690918,
      "reward_std": 0.07426881790161133,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6011915057897568,
      "step": 928
    },
    {
      "completion_length": 95.65625,
      "epoch": 0.5989684074790458,
      "grad_norm": 14.408288955688477,
      "kl": 0.15185546875,
      "learning_rate": 7.005157962604771e-07,
      "loss": 0.0061,
      "reward": 1.5045355558395386,
      "reward_std": 0.05852143466472626,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5045354962348938,
      "step": 929
    },
    {
      "completion_length": 111.78125,
      "epoch": 0.5996131528046421,
      "grad_norm": 14.51344108581543,
      "kl": 0.13134765625,
      "learning_rate": 7.001934235976789e-07,
      "loss": 0.0053,
      "reward": 1.6496663093566895,
      "reward_std": 0.07880590856075287,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6496663391590118,
      "step": 930
    },
    {
      "completion_length": 96.875,
      "epoch": 0.6002578981302386,
      "grad_norm": 27.438138961791992,
      "kl": 0.1298828125,
      "learning_rate": 6.998710509348806e-07,
      "loss": 0.0052,
      "reward": 1.567421317100525,
      "reward_std": 0.058061445131897926,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5674213171005249,
      "step": 931
    },
    {
      "completion_length": 99.703125,
      "epoch": 0.600902643455835,
      "grad_norm": 12.056415557861328,
      "kl": 0.126708984375,
      "learning_rate": 6.995486782720826e-07,
      "loss": 0.0051,
      "reward": 1.5093879699707031,
      "reward_std": 0.09708208404481411,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.509387880563736,
      "step": 932
    },
    {
      "completion_length": 92.359375,
      "epoch": 0.6015473887814313,
      "grad_norm": 32.935951232910156,
      "kl": 0.15283203125,
      "learning_rate": 6.992263056092843e-07,
      "loss": 0.0061,
      "reward": 1.6477266550064087,
      "reward_std": 0.11407394707202911,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6477266252040863,
      "step": 933
    },
    {
      "completion_length": 96.328125,
      "epoch": 0.6021921341070278,
      "grad_norm": 7.755570888519287,
      "kl": 0.143798828125,
      "learning_rate": 6.989039329464861e-07,
      "loss": 0.0058,
      "reward": 1.8171811699867249,
      "reward_std": 0.12367031536996365,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.8171811997890472,
      "step": 934
    },
    {
      "completion_length": 93.75,
      "epoch": 0.6028368794326241,
      "grad_norm": 25.14842987060547,
      "kl": 0.124755859375,
      "learning_rate": 6.985815602836878e-07,
      "loss": 0.005,
      "reward": 1.349519431591034,
      "reward_std": 0.05427607707679272,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3495194613933563,
      "step": 935
    },
    {
      "completion_length": 93.921875,
      "epoch": 0.6034816247582205,
      "grad_norm": 56.53002166748047,
      "kl": 0.1259765625,
      "learning_rate": 6.982591876208898e-07,
      "loss": 0.005,
      "reward": 1.508469820022583,
      "reward_std": 0.04055575653910637,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5084697008132935,
      "step": 936
    },
    {
      "completion_length": 96.390625,
      "epoch": 0.6041263700838169,
      "grad_norm": 18.896730422973633,
      "kl": 0.14501953125,
      "learning_rate": 6.979368149580915e-07,
      "loss": 0.0058,
      "reward": 1.5613884329795837,
      "reward_std": 0.07465831749141216,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5613884329795837,
      "step": 937
    },
    {
      "completion_length": 94.53125,
      "epoch": 0.6047711154094133,
      "grad_norm": 28.830486297607422,
      "kl": 0.1220703125,
      "learning_rate": 6.976144422952933e-07,
      "loss": 0.0049,
      "reward": 1.5561841130256653,
      "reward_std": 0.05068258661776781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5561841726303101,
      "step": 938
    },
    {
      "completion_length": 89.59375,
      "epoch": 0.6054158607350096,
      "grad_norm": 20.040565490722656,
      "kl": 0.15966796875,
      "learning_rate": 6.972920696324951e-07,
      "loss": 0.0064,
      "reward": 1.5743331909179688,
      "reward_std": 0.07431630790233612,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5743332505226135,
      "step": 939
    },
    {
      "completion_length": 94.390625,
      "epoch": 0.6060606060606061,
      "grad_norm": 16.654029846191406,
      "kl": 0.16015625,
      "learning_rate": 6.96969696969697e-07,
      "loss": 0.0064,
      "reward": 1.7051098346710205,
      "reward_std": 0.1160728745162487,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7051098048686981,
      "step": 940
    },
    {
      "completion_length": 96.15625,
      "epoch": 0.6067053513862024,
      "grad_norm": 17.097835540771484,
      "kl": 0.167236328125,
      "learning_rate": 6.966473243068987e-07,
      "loss": 0.0067,
      "reward": 1.5818954706192017,
      "reward_std": 0.09519368968904018,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5818954259157181,
      "step": 941
    },
    {
      "completion_length": 90.3125,
      "epoch": 0.6073500967117988,
      "grad_norm": 17.13744354248047,
      "kl": 0.11962890625,
      "learning_rate": 6.963249516441006e-07,
      "loss": 0.0048,
      "reward": 1.6105692386627197,
      "reward_std": 0.11062491312623024,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6105693280696869,
      "step": 942
    },
    {
      "completion_length": 89.703125,
      "epoch": 0.6079948420373952,
      "grad_norm": 41.86802291870117,
      "kl": 0.1533203125,
      "learning_rate": 6.960025789813023e-07,
      "loss": 0.0061,
      "reward": 1.5191384553909302,
      "reward_std": 0.05962220951914787,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5191383957862854,
      "step": 943
    },
    {
      "completion_length": 99.4375,
      "epoch": 0.6086395873629916,
      "grad_norm": 53.935115814208984,
      "kl": 0.12451171875,
      "learning_rate": 6.956802063185042e-07,
      "loss": 0.005,
      "reward": 1.477087378501892,
      "reward_std": 0.07249218598008156,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4770873636007309,
      "step": 944
    },
    {
      "completion_length": 89.984375,
      "epoch": 0.6092843326885881,
      "grad_norm": 23.599111557006836,
      "kl": 0.13818359375,
      "learning_rate": 6.95357833655706e-07,
      "loss": 0.0055,
      "reward": 1.7460429668426514,
      "reward_std": 0.07097552344202995,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7460429072380066,
      "step": 945
    },
    {
      "completion_length": 92.765625,
      "epoch": 0.6099290780141844,
      "grad_norm": 18.278663635253906,
      "kl": 0.135498046875,
      "learning_rate": 6.950354609929078e-07,
      "loss": 0.0054,
      "reward": 1.7038219571113586,
      "reward_std": 0.07454410195350647,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.703821986913681,
      "step": 946
    },
    {
      "completion_length": 87.125,
      "epoch": 0.6105738233397808,
      "grad_norm": 25.546600341796875,
      "kl": 0.1328125,
      "learning_rate": 6.947130883301095e-07,
      "loss": 0.0053,
      "reward": 1.6144275069236755,
      "reward_std": 0.06358279846608639,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6144274771213531,
      "step": 947
    },
    {
      "completion_length": 89.40625,
      "epoch": 0.6112185686653772,
      "grad_norm": 23.264015197753906,
      "kl": 0.19189453125,
      "learning_rate": 6.943907156673114e-07,
      "loss": 0.0077,
      "reward": 1.5967767834663391,
      "reward_std": 0.14299481362104416,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5967767834663391,
      "step": 948
    },
    {
      "completion_length": 85.84375,
      "epoch": 0.6118633139909736,
      "grad_norm": 68.70216369628906,
      "kl": 0.12109375,
      "learning_rate": 6.940683430045132e-07,
      "loss": 0.0048,
      "reward": 1.452661395072937,
      "reward_std": 0.050826674327254295,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45266131311655045,
      "step": 949
    },
    {
      "completion_length": 91.140625,
      "epoch": 0.6125080593165699,
      "grad_norm": 20.137420654296875,
      "kl": 0.1220703125,
      "learning_rate": 6.93745970341715e-07,
      "loss": 0.0049,
      "reward": 1.5065082907676697,
      "reward_std": 0.10698516108095646,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5065082460641861,
      "step": 950
    },
    {
      "completion_length": 85.6875,
      "epoch": 0.6131528046421664,
      "grad_norm": 14.560029983520508,
      "kl": 0.1396484375,
      "learning_rate": 6.934235976789168e-07,
      "loss": 0.0056,
      "reward": 1.513734221458435,
      "reward_std": 0.06140206754207611,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5137341916561127,
      "step": 951
    },
    {
      "completion_length": 84.484375,
      "epoch": 0.6137975499677627,
      "grad_norm": 34.63451385498047,
      "kl": 0.1396484375,
      "learning_rate": 6.931012250161186e-07,
      "loss": 0.0056,
      "reward": 1.5841580629348755,
      "reward_std": 0.0958426408469677,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5841581225395203,
      "step": 952
    },
    {
      "completion_length": 88.15625,
      "epoch": 0.6144422952933591,
      "grad_norm": 17.28553009033203,
      "kl": 0.13916015625,
      "learning_rate": 6.927788523533204e-07,
      "loss": 0.0056,
      "reward": 1.4624040126800537,
      "reward_std": 0.06560158357024193,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46240395307540894,
      "step": 953
    },
    {
      "completion_length": 91.765625,
      "epoch": 0.6150870406189555,
      "grad_norm": 18.49038314819336,
      "kl": 0.14013671875,
      "learning_rate": 6.924564796905223e-07,
      "loss": 0.0056,
      "reward": 1.7362635135650635,
      "reward_std": 0.05572156980633736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7362635731697083,
      "step": 954
    },
    {
      "completion_length": 99.15625,
      "epoch": 0.6157317859445519,
      "grad_norm": 13.38481616973877,
      "kl": 0.116943359375,
      "learning_rate": 6.92134107027724e-07,
      "loss": 0.0047,
      "reward": 1.662828803062439,
      "reward_std": 0.09047294408082962,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6628288626670837,
      "step": 955
    },
    {
      "completion_length": 87.9375,
      "epoch": 0.6163765312701482,
      "grad_norm": 11.720268249511719,
      "kl": 0.13134765625,
      "learning_rate": 6.918117343649258e-07,
      "loss": 0.0053,
      "reward": 1.7224065065383911,
      "reward_std": 0.12152724154293537,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7380314767360687,
      "step": 956
    },
    {
      "completion_length": 89.796875,
      "epoch": 0.6170212765957447,
      "grad_norm": 51.41786193847656,
      "kl": 0.12744140625,
      "learning_rate": 6.914893617021277e-07,
      "loss": 0.0051,
      "reward": 1.600498080253601,
      "reward_std": 0.0731177031993866,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6004980802536011,
      "step": 957
    },
    {
      "completion_length": 80.34375,
      "epoch": 0.6176660219213411,
      "grad_norm": 21.15074920654297,
      "kl": 0.1494140625,
      "learning_rate": 6.911669890393295e-07,
      "loss": 0.006,
      "reward": 1.5226842164993286,
      "reward_std": 0.15971365571022034,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5226842164993286,
      "step": 958
    },
    {
      "completion_length": 88.546875,
      "epoch": 0.6183107672469375,
      "grad_norm": 20.88332176208496,
      "kl": 0.132080078125,
      "learning_rate": 6.908446163765312e-07,
      "loss": 0.0053,
      "reward": 1.5318822264671326,
      "reward_std": 0.08424959145486355,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5318822264671326,
      "step": 959
    },
    {
      "completion_length": 84.0,
      "epoch": 0.6189555125725339,
      "grad_norm": 12.1968994140625,
      "kl": 0.16162109375,
      "learning_rate": 6.90522243713733e-07,
      "loss": 0.0065,
      "reward": 1.609505832195282,
      "reward_std": 0.1656803898513317,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.625130832195282,
      "step": 960
    },
    {
      "completion_length": 88.703125,
      "epoch": 0.6196002578981302,
      "grad_norm": 37.65996551513672,
      "kl": 0.138916015625,
      "learning_rate": 6.901998710509348e-07,
      "loss": 0.0056,
      "reward": 1.6943501234054565,
      "reward_std": 0.10790320858359337,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6943500936031342,
      "step": 961
    },
    {
      "completion_length": 89.828125,
      "epoch": 0.6202450032237267,
      "grad_norm": 14.622736930847168,
      "kl": 0.130859375,
      "learning_rate": 6.898774983881367e-07,
      "loss": 0.0052,
      "reward": 1.6017565727233887,
      "reward_std": 0.09940549731254578,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6017565131187439,
      "step": 962
    },
    {
      "completion_length": 91.453125,
      "epoch": 0.620889748549323,
      "grad_norm": 21.332942962646484,
      "kl": 0.1259765625,
      "learning_rate": 6.895551257253384e-07,
      "loss": 0.005,
      "reward": 1.3970187306404114,
      "reward_std": 0.0625982228666544,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3970186710357666,
      "step": 963
    },
    {
      "completion_length": 81.546875,
      "epoch": 0.6215344938749194,
      "grad_norm": 25.18134117126465,
      "kl": 0.18994140625,
      "learning_rate": 6.892327530625403e-07,
      "loss": 0.0076,
      "reward": 1.5218143463134766,
      "reward_std": 0.19736109673976898,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5374393910169601,
      "step": 964
    },
    {
      "completion_length": 81.8125,
      "epoch": 0.6221792392005158,
      "grad_norm": 13.722188949584961,
      "kl": 0.15478515625,
      "learning_rate": 6.88910380399742e-07,
      "loss": 0.0062,
      "reward": 1.4993084073066711,
      "reward_std": 0.06540745310485363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49930842220783234,
      "step": 965
    },
    {
      "completion_length": 86.5,
      "epoch": 0.6228239845261122,
      "grad_norm": 11.098575592041016,
      "kl": 0.14697265625,
      "learning_rate": 6.885880077369439e-07,
      "loss": 0.0059,
      "reward": 1.5485892295837402,
      "reward_std": 0.07327968627214432,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5485892295837402,
      "step": 966
    },
    {
      "completion_length": 104.078125,
      "epoch": 0.6234687298517085,
      "grad_norm": 18.58123016357422,
      "kl": 0.130859375,
      "learning_rate": 6.882656350741457e-07,
      "loss": 0.0052,
      "reward": 1.6219686269760132,
      "reward_std": 0.07606706768274307,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6219685971736908,
      "step": 967
    },
    {
      "completion_length": 90.3125,
      "epoch": 0.624113475177305,
      "grad_norm": 16.83329200744629,
      "kl": 0.15576171875,
      "learning_rate": 6.879432624113475e-07,
      "loss": 0.0062,
      "reward": 1.4810823202133179,
      "reward_std": 0.06910162046551704,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4810822904109955,
      "step": 968
    },
    {
      "completion_length": 86.9375,
      "epoch": 0.6247582205029013,
      "grad_norm": 14.21639633178711,
      "kl": 0.111083984375,
      "learning_rate": 6.876208897485492e-07,
      "loss": 0.0044,
      "reward": 1.6480799913406372,
      "reward_std": 0.07823758944869041,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6480799615383148,
      "step": 969
    },
    {
      "completion_length": 86.96875,
      "epoch": 0.6254029658284977,
      "grad_norm": 27.79960823059082,
      "kl": 0.137451171875,
      "learning_rate": 6.872985170857512e-07,
      "loss": 0.0055,
      "reward": 1.760128676891327,
      "reward_std": 0.07539523765444756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7601286768913269,
      "step": 970
    },
    {
      "completion_length": 98.34375,
      "epoch": 0.6260477111540941,
      "grad_norm": 156.4920654296875,
      "kl": 0.1728515625,
      "learning_rate": 6.869761444229529e-07,
      "loss": 0.0069,
      "reward": 1.3526663184165955,
      "reward_std": 0.0894409529864788,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35266634821891785,
      "step": 971
    },
    {
      "completion_length": 90.359375,
      "epoch": 0.6266924564796905,
      "grad_norm": 23.380796432495117,
      "kl": 0.133544921875,
      "learning_rate": 6.866537717601547e-07,
      "loss": 0.0054,
      "reward": 1.6337125301361084,
      "reward_std": 0.13482321798801422,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6337125897407532,
      "step": 972
    },
    {
      "completion_length": 99.96875,
      "epoch": 0.627337201805287,
      "grad_norm": 147.32347106933594,
      "kl": 0.1103515625,
      "learning_rate": 6.863313990973565e-07,
      "loss": 0.0044,
      "reward": 1.5924115180969238,
      "reward_std": 0.10236813127994537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5924114882946014,
      "step": 973
    },
    {
      "completion_length": 97.453125,
      "epoch": 0.6279819471308833,
      "grad_norm": 15.938390731811523,
      "kl": 0.1357421875,
      "learning_rate": 6.860090264345583e-07,
      "loss": 0.0054,
      "reward": 1.6408860683441162,
      "reward_std": 0.13076971843838692,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6565110683441162,
      "step": 974
    },
    {
      "completion_length": 97.0625,
      "epoch": 0.6286266924564797,
      "grad_norm": 29.58421516418457,
      "kl": 0.104736328125,
      "learning_rate": 6.856866537717601e-07,
      "loss": 0.0042,
      "reward": 1.6835073232650757,
      "reward_std": 0.0805157907307148,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6835072636604309,
      "step": 975
    },
    {
      "completion_length": 101.703125,
      "epoch": 0.6292714377820761,
      "grad_norm": 11.460612297058105,
      "kl": 0.13037109375,
      "learning_rate": 6.85364281108962e-07,
      "loss": 0.0052,
      "reward": 1.4942415356636047,
      "reward_std": 0.0735623836517334,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49424150586128235,
      "step": 976
    },
    {
      "completion_length": 97.609375,
      "epoch": 0.6299161831076725,
      "grad_norm": 54.10033416748047,
      "kl": 0.1171875,
      "learning_rate": 6.850419084461637e-07,
      "loss": 0.0047,
      "reward": 1.5240637063980103,
      "reward_std": 0.06277331709861755,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5240637063980103,
      "step": 977
    },
    {
      "completion_length": 98.9375,
      "epoch": 0.6305609284332688,
      "grad_norm": 23.348764419555664,
      "kl": 0.109619140625,
      "learning_rate": 6.847195357833655e-07,
      "loss": 0.0044,
      "reward": 1.5499730706214905,
      "reward_std": 0.11529890447854996,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5499730408191681,
      "step": 978
    },
    {
      "completion_length": 101.265625,
      "epoch": 0.6312056737588653,
      "grad_norm": 36.193199157714844,
      "kl": 0.10595703125,
      "learning_rate": 6.843971631205674e-07,
      "loss": 0.0042,
      "reward": 1.7043808102607727,
      "reward_std": 0.065880686044693,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7043807804584503,
      "step": 979
    },
    {
      "completion_length": 103.21875,
      "epoch": 0.6318504190844616,
      "grad_norm": 25.998506546020508,
      "kl": 0.113037109375,
      "learning_rate": 6.840747904577692e-07,
      "loss": 0.0045,
      "reward": 1.5761331915855408,
      "reward_std": 0.07709749229252338,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5761331915855408,
      "step": 980
    },
    {
      "completion_length": 101.15625,
      "epoch": 0.632495164410058,
      "grad_norm": 25.730180740356445,
      "kl": 0.130615234375,
      "learning_rate": 6.837524177949709e-07,
      "loss": 0.0052,
      "reward": 1.3551477193832397,
      "reward_std": 0.13733485713601112,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35514768958091736,
      "step": 981
    },
    {
      "completion_length": 105.4375,
      "epoch": 0.6331399097356544,
      "grad_norm": 29.705520629882812,
      "kl": 0.12841796875,
      "learning_rate": 6.834300451321727e-07,
      "loss": 0.0051,
      "reward": 1.4724050164222717,
      "reward_std": 0.09875421039760113,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4724050760269165,
      "step": 982
    },
    {
      "completion_length": 109.0625,
      "epoch": 0.6337846550612508,
      "grad_norm": 21.819368362426758,
      "kl": 0.130859375,
      "learning_rate": 6.831076724693746e-07,
      "loss": 0.0052,
      "reward": 1.5190939903259277,
      "reward_std": 0.18469242006540298,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.534718930721283,
      "step": 983
    },
    {
      "completion_length": 91.359375,
      "epoch": 0.6344294003868471,
      "grad_norm": 15.404848098754883,
      "kl": 0.123779296875,
      "learning_rate": 6.827852998065764e-07,
      "loss": 0.0049,
      "reward": 1.410255491733551,
      "reward_std": 0.13094928860664368,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41025547683238983,
      "step": 984
    },
    {
      "completion_length": 102.015625,
      "epoch": 0.6350741457124436,
      "grad_norm": 25.152095794677734,
      "kl": 0.12158203125,
      "learning_rate": 6.824629271437781e-07,
      "loss": 0.0049,
      "reward": 1.477942705154419,
      "reward_std": 0.06228645518422127,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47794269025325775,
      "step": 985
    },
    {
      "completion_length": 103.4375,
      "epoch": 0.63571889103804,
      "grad_norm": 16.888469696044922,
      "kl": 0.123291015625,
      "learning_rate": 6.8214055448098e-07,
      "loss": 0.0049,
      "reward": 1.4599791765213013,
      "reward_std": 0.05538452975451946,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4599791467189789,
      "step": 986
    },
    {
      "completion_length": 108.40625,
      "epoch": 0.6363636363636364,
      "grad_norm": 11.82778549194336,
      "kl": 0.119140625,
      "learning_rate": 6.818181818181817e-07,
      "loss": 0.0048,
      "reward": 1.5686384439468384,
      "reward_std": 0.1496180146932602,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5686384737491608,
      "step": 987
    },
    {
      "completion_length": 101.546875,
      "epoch": 0.6370083816892328,
      "grad_norm": 7.283498287200928,
      "kl": 0.110107421875,
      "learning_rate": 6.814958091553836e-07,
      "loss": 0.0044,
      "reward": 1.6922195553779602,
      "reward_std": 0.1970614790916443,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7078445553779602,
      "step": 988
    },
    {
      "completion_length": 93.859375,
      "epoch": 0.6376531270148291,
      "grad_norm": 13.84801197052002,
      "kl": 0.14697265625,
      "learning_rate": 6.811734364925854e-07,
      "loss": 0.0059,
      "reward": 1.6163573861122131,
      "reward_std": 0.1375247687101364,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6163573861122131,
      "step": 989
    },
    {
      "completion_length": 107.140625,
      "epoch": 0.6382978723404256,
      "grad_norm": 26.76303482055664,
      "kl": 0.24169921875,
      "learning_rate": 6.808510638297872e-07,
      "loss": 0.0097,
      "reward": 1.4439324140548706,
      "reward_std": 0.09625233709812164,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4595574140548706,
      "step": 990
    },
    {
      "completion_length": 101.75,
      "epoch": 0.6389426176660219,
      "grad_norm": 23.95302391052246,
      "kl": 0.12158203125,
      "learning_rate": 6.805286911669889e-07,
      "loss": 0.0049,
      "reward": 1.6145684123039246,
      "reward_std": 0.0981244370341301,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6145684719085693,
      "step": 991
    },
    {
      "completion_length": 104.6875,
      "epoch": 0.6395873629916183,
      "grad_norm": 19.830524444580078,
      "kl": 0.107421875,
      "learning_rate": 6.802063185041909e-07,
      "loss": 0.0043,
      "reward": 1.4319545030593872,
      "reward_std": 0.06444331258535385,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4319545179605484,
      "step": 992
    },
    {
      "completion_length": 109.578125,
      "epoch": 0.6402321083172147,
      "grad_norm": 12.206172943115234,
      "kl": 0.13330078125,
      "learning_rate": 6.798839458413926e-07,
      "loss": 0.0053,
      "reward": 1.7099278569221497,
      "reward_std": 0.08155185729265213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7099278569221497,
      "step": 993
    },
    {
      "completion_length": 115.390625,
      "epoch": 0.6408768536428111,
      "grad_norm": 25.8281192779541,
      "kl": 0.12841796875,
      "learning_rate": 6.795615731785944e-07,
      "loss": 0.0051,
      "reward": 1.6406802535057068,
      "reward_std": 0.11553433537483215,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6406802237033844,
      "step": 994
    },
    {
      "completion_length": 106.21875,
      "epoch": 0.6415215989684074,
      "grad_norm": 13.230984687805176,
      "kl": 0.11474609375,
      "learning_rate": 6.792392005157962e-07,
      "loss": 0.0046,
      "reward": 1.4972352981567383,
      "reward_std": 0.08803327940404415,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4972353130578995,
      "step": 995
    },
    {
      "completion_length": 104.953125,
      "epoch": 0.6421663442940039,
      "grad_norm": 10.515935897827148,
      "kl": 0.107177734375,
      "learning_rate": 6.789168278529981e-07,
      "loss": 0.0043,
      "reward": 1.7526929378509521,
      "reward_std": 0.0644787959754467,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7526930272579193,
      "step": 996
    },
    {
      "completion_length": 105.40625,
      "epoch": 0.6428110896196002,
      "grad_norm": 18.340421676635742,
      "kl": 0.1064453125,
      "learning_rate": 6.785944551901998e-07,
      "loss": 0.0043,
      "reward": 1.5969550013542175,
      "reward_std": 0.11528073251247406,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5969549715518951,
      "step": 997
    },
    {
      "completion_length": 107.6875,
      "epoch": 0.6434558349451966,
      "grad_norm": 15.871825218200684,
      "kl": 0.113525390625,
      "learning_rate": 6.782720825274017e-07,
      "loss": 0.0045,
      "reward": 1.522782325744629,
      "reward_std": 0.11090462654829025,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5227823406457901,
      "step": 998
    },
    {
      "completion_length": 115.90625,
      "epoch": 0.6441005802707931,
      "grad_norm": 23.02735710144043,
      "kl": 0.182861328125,
      "learning_rate": 6.779497098646034e-07,
      "loss": 0.0073,
      "reward": 1.3356384634971619,
      "reward_std": 0.039747599977999926,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33563845232129097,
      "step": 999
    },
    {
      "completion_length": 112.265625,
      "epoch": 0.6447453255963894,
      "grad_norm": 29.998191833496094,
      "kl": 0.11181640625,
      "learning_rate": 6.776273372018052e-07,
      "loss": 0.0045,
      "reward": 1.5243074893951416,
      "reward_std": 0.07966102100908756,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.52430759370327,
      "step": 1000
    },
    {
      "completion_length": 101.875,
      "epoch": 0.6453900709219859,
      "grad_norm": 13.788435935974121,
      "kl": 0.11328125,
      "learning_rate": 6.77304964539007e-07,
      "loss": 0.0045,
      "reward": 1.5019648671150208,
      "reward_std": 0.064224723726511,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5019648671150208,
      "step": 1001
    },
    {
      "completion_length": 98.0625,
      "epoch": 0.6460348162475822,
      "grad_norm": 22.711002349853516,
      "kl": 0.13671875,
      "learning_rate": 6.769825918762089e-07,
      "loss": 0.0055,
      "reward": 1.465948462486267,
      "reward_std": 0.05411083064973354,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4659484773874283,
      "step": 1002
    },
    {
      "completion_length": 104.703125,
      "epoch": 0.6466795615731786,
      "grad_norm": 22.697656631469727,
      "kl": 0.13720703125,
      "learning_rate": 6.766602192134106e-07,
      "loss": 0.0055,
      "reward": 1.5236431956291199,
      "reward_std": 0.1108100637793541,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5236431360244751,
      "step": 1003
    },
    {
      "completion_length": 113.78125,
      "epoch": 0.647324306898775,
      "grad_norm": 13.408069610595703,
      "kl": 0.122314453125,
      "learning_rate": 6.763378465506124e-07,
      "loss": 0.0049,
      "reward": 1.5940637588500977,
      "reward_std": 0.09507790580391884,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5940638333559036,
      "step": 1004
    },
    {
      "completion_length": 96.828125,
      "epoch": 0.6479690522243714,
      "grad_norm": 34.19378662109375,
      "kl": 0.1181640625,
      "learning_rate": 6.760154738878143e-07,
      "loss": 0.0047,
      "reward": 1.6176080703735352,
      "reward_std": 0.13371532410383224,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6176081299781799,
      "step": 1005
    },
    {
      "completion_length": 107.453125,
      "epoch": 0.6486137975499677,
      "grad_norm": 27.635814666748047,
      "kl": 0.10400390625,
      "learning_rate": 6.756931012250161e-07,
      "loss": 0.0042,
      "reward": 1.4466070532798767,
      "reward_std": 0.1371947005391121,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4466071128845215,
      "step": 1006
    },
    {
      "completion_length": 108.6875,
      "epoch": 0.6492585428755642,
      "grad_norm": 9.812318801879883,
      "kl": 0.1337890625,
      "learning_rate": 6.753707285622178e-07,
      "loss": 0.0053,
      "reward": 1.5490465760231018,
      "reward_std": 0.08941208943724632,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5490465462207794,
      "step": 1007
    },
    {
      "completion_length": 113.84375,
      "epoch": 0.6499032882011605,
      "grad_norm": 15.088581085205078,
      "kl": 0.13232421875,
      "learning_rate": 6.750483558994197e-07,
      "loss": 0.0053,
      "reward": 1.6374335885047913,
      "reward_std": 0.13946302980184555,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6530585885047913,
      "step": 1008
    },
    {
      "completion_length": 100.46875,
      "epoch": 0.6505480335267569,
      "grad_norm": 18.111522674560547,
      "kl": 0.19873046875,
      "learning_rate": 6.747259832366215e-07,
      "loss": 0.008,
      "reward": 1.5997966527938843,
      "reward_std": 0.18083975464105606,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.615421712398529,
      "step": 1009
    },
    {
      "completion_length": 119.78125,
      "epoch": 0.6511927788523533,
      "grad_norm": 10.134917259216309,
      "kl": 0.1083984375,
      "learning_rate": 6.744036105738233e-07,
      "loss": 0.0043,
      "reward": 1.6330845355987549,
      "reward_std": 0.08692387118935585,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6330845654010773,
      "step": 1010
    },
    {
      "completion_length": 105.765625,
      "epoch": 0.6518375241779497,
      "grad_norm": 24.195541381835938,
      "kl": 0.11962890625,
      "learning_rate": 6.740812379110251e-07,
      "loss": 0.0048,
      "reward": 1.5731955170631409,
      "reward_std": 0.06743450090289116,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5731955468654633,
      "step": 1011
    },
    {
      "completion_length": 101.953125,
      "epoch": 0.6524822695035462,
      "grad_norm": 63.89750289916992,
      "kl": 0.120849609375,
      "learning_rate": 6.737588652482269e-07,
      "loss": 0.0048,
      "reward": 1.4986768960952759,
      "reward_std": 0.10133867338299751,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4986769109964371,
      "step": 1012
    },
    {
      "completion_length": 96.9375,
      "epoch": 0.6531270148291425,
      "grad_norm": 10.6175537109375,
      "kl": 0.1181640625,
      "learning_rate": 6.734364925854286e-07,
      "loss": 0.0047,
      "reward": 1.6250901818275452,
      "reward_std": 0.07033353298902512,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6250901520252228,
      "step": 1013
    },
    {
      "completion_length": 95.875,
      "epoch": 0.6537717601547389,
      "grad_norm": 113.03011322021484,
      "kl": 0.1142578125,
      "learning_rate": 6.731141199226306e-07,
      "loss": 0.0046,
      "reward": 1.587643027305603,
      "reward_std": 0.12430031597614288,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.587643027305603,
      "step": 1014
    },
    {
      "completion_length": 94.421875,
      "epoch": 0.6544165054803353,
      "grad_norm": 16.419910430908203,
      "kl": 0.110595703125,
      "learning_rate": 6.727917472598323e-07,
      "loss": 0.0044,
      "reward": 1.5937945246696472,
      "reward_std": 0.10029425844550133,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5937945246696472,
      "step": 1015
    },
    {
      "completion_length": 101.0625,
      "epoch": 0.6550612508059317,
      "grad_norm": 23.794921875,
      "kl": 0.134765625,
      "learning_rate": 6.724693745970341e-07,
      "loss": 0.0054,
      "reward": 1.6735421419143677,
      "reward_std": 0.10969025641679764,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6735422015190125,
      "step": 1016
    },
    {
      "completion_length": 110.609375,
      "epoch": 0.655705996131528,
      "grad_norm": 12.621672630310059,
      "kl": 0.11376953125,
      "learning_rate": 6.721470019342359e-07,
      "loss": 0.0045,
      "reward": 1.71983140707016,
      "reward_std": 0.06134466826915741,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7198314368724823,
      "step": 1017
    },
    {
      "completion_length": 98.25,
      "epoch": 0.6563507414571245,
      "grad_norm": 41.569644927978516,
      "kl": 0.117431640625,
      "learning_rate": 6.718246292714378e-07,
      "loss": 0.0047,
      "reward": 1.5260882377624512,
      "reward_std": 0.09492732584476471,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5260882377624512,
      "step": 1018
    },
    {
      "completion_length": 100.390625,
      "epoch": 0.6569954867827208,
      "grad_norm": 15.814488410949707,
      "kl": 0.169921875,
      "learning_rate": 6.715022566086395e-07,
      "loss": 0.0068,
      "reward": 1.5511394739151,
      "reward_std": 0.10130210220813751,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5511394739151001,
      "step": 1019
    },
    {
      "completion_length": 93.1875,
      "epoch": 0.6576402321083172,
      "grad_norm": 25.868892669677734,
      "kl": 0.1103515625,
      "learning_rate": 6.711798839458414e-07,
      "loss": 0.0044,
      "reward": 1.7657926678657532,
      "reward_std": 0.07120348140597343,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.765792727470398,
      "step": 1020
    },
    {
      "completion_length": 105.140625,
      "epoch": 0.6582849774339136,
      "grad_norm": 11.2671537399292,
      "kl": 0.1015625,
      "learning_rate": 6.708575112830431e-07,
      "loss": 0.0041,
      "reward": 1.5160332322120667,
      "reward_std": 0.06816467270255089,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5160332322120667,
      "step": 1021
    },
    {
      "completion_length": 103.75,
      "epoch": 0.65892972275951,
      "grad_norm": 11.783982276916504,
      "kl": 0.10400390625,
      "learning_rate": 6.70535138620245e-07,
      "loss": 0.0042,
      "reward": 1.634143590927124,
      "reward_std": 0.0808180458843708,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6341435015201569,
      "step": 1022
    },
    {
      "completion_length": 109.6875,
      "epoch": 0.6595744680851063,
      "grad_norm": 18.366363525390625,
      "kl": 0.1240234375,
      "learning_rate": 6.702127659574468e-07,
      "loss": 0.005,
      "reward": 1.7418678998947144,
      "reward_std": 0.09646868705749512,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7418678402900696,
      "step": 1023
    },
    {
      "completion_length": 104.390625,
      "epoch": 0.6602192134107028,
      "grad_norm": 44.52997589111328,
      "kl": 0.12939453125,
      "learning_rate": 6.698903932946486e-07,
      "loss": 0.0052,
      "reward": 1.4403961896896362,
      "reward_std": 0.13813015818595886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.440396249294281,
      "step": 1024
    },
    {
      "completion_length": 101.8125,
      "epoch": 0.6608639587362991,
      "grad_norm": 13.398497581481934,
      "kl": 0.12158203125,
      "learning_rate": 6.695680206318503e-07,
      "loss": 0.0049,
      "reward": 1.6842644214630127,
      "reward_std": 0.12250592187047005,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6842643916606903,
      "step": 1025
    },
    {
      "completion_length": 111.1875,
      "epoch": 0.6615087040618955,
      "grad_norm": 37.01210403442383,
      "kl": 0.12744140625,
      "learning_rate": 6.692456479690521e-07,
      "loss": 0.0051,
      "reward": 1.5435105562210083,
      "reward_std": 0.08950828574597836,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5435105115175247,
      "step": 1026
    },
    {
      "completion_length": 109.078125,
      "epoch": 0.662153449387492,
      "grad_norm": 38.497901916503906,
      "kl": 0.12353515625,
      "learning_rate": 6.68923275306254e-07,
      "loss": 0.0049,
      "reward": 1.4949541687965393,
      "reward_std": 0.10163058340549469,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4949541985988617,
      "step": 1027
    },
    {
      "completion_length": 102.640625,
      "epoch": 0.6627981947130883,
      "grad_norm": 18.939176559448242,
      "kl": 0.11474609375,
      "learning_rate": 6.686009026434558e-07,
      "loss": 0.0046,
      "reward": 1.5689424276351929,
      "reward_std": 0.0853864960372448,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5689424276351929,
      "step": 1028
    },
    {
      "completion_length": 111.734375,
      "epoch": 0.6634429400386848,
      "grad_norm": 8.950363159179688,
      "kl": 0.105224609375,
      "learning_rate": 6.682785299806575e-07,
      "loss": 0.0042,
      "reward": 1.5498056411743164,
      "reward_std": 0.05270315520465374,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5498056858778,
      "step": 1029
    },
    {
      "completion_length": 103.359375,
      "epoch": 0.6640876853642811,
      "grad_norm": 9.672654151916504,
      "kl": 0.1142578125,
      "learning_rate": 6.679561573178594e-07,
      "loss": 0.0046,
      "reward": 1.630305826663971,
      "reward_std": 0.04982496798038483,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6303058564662933,
      "step": 1030
    },
    {
      "completion_length": 103.3125,
      "epoch": 0.6647324306898775,
      "grad_norm": 37.13180923461914,
      "kl": 0.103515625,
      "learning_rate": 6.676337846550613e-07,
      "loss": 0.0041,
      "reward": 1.6453443765640259,
      "reward_std": 0.09188783913850784,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6453443467617035,
      "step": 1031
    },
    {
      "completion_length": 98.640625,
      "epoch": 0.6653771760154739,
      "grad_norm": 13.750219345092773,
      "kl": 0.128173828125,
      "learning_rate": 6.67311411992263e-07,
      "loss": 0.0051,
      "reward": 1.4959722757339478,
      "reward_std": 0.07575977221131325,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49597224593162537,
      "step": 1032
    },
    {
      "completion_length": 116.953125,
      "epoch": 0.6660219213410703,
      "grad_norm": 18.885990142822266,
      "kl": 0.1201171875,
      "learning_rate": 6.669890393294648e-07,
      "loss": 0.0048,
      "reward": 1.507498860359192,
      "reward_std": 0.09108571708202362,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5074988901615143,
      "step": 1033
    },
    {
      "completion_length": 106.109375,
      "epoch": 0.6666666666666666,
      "grad_norm": 13.014422416687012,
      "kl": 0.1064453125,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.0043,
      "reward": 1.7497589588165283,
      "reward_std": 0.04258188419044018,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7497589588165283,
      "step": 1034
    },
    {
      "completion_length": 88.640625,
      "epoch": 0.6673114119922631,
      "grad_norm": 12.072609901428223,
      "kl": 0.14990234375,
      "learning_rate": 6.663442940038685e-07,
      "loss": 0.006,
      "reward": 1.4128879308700562,
      "reward_std": 0.05543842911720276,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41288794577121735,
      "step": 1035
    },
    {
      "completion_length": 95.421875,
      "epoch": 0.6679561573178594,
      "grad_norm": 25.26651954650879,
      "kl": 0.1201171875,
      "learning_rate": 6.660219213410703e-07,
      "loss": 0.0048,
      "reward": 1.6416820287704468,
      "reward_std": 0.10006970167160034,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6416819393634796,
      "step": 1036
    },
    {
      "completion_length": 109.75,
      "epoch": 0.6686009026434558,
      "grad_norm": 35.28297424316406,
      "kl": 0.114501953125,
      "learning_rate": 6.656995486782721e-07,
      "loss": 0.0046,
      "reward": 1.48683100938797,
      "reward_std": 0.12712857872247696,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48683103919029236,
      "step": 1037
    },
    {
      "completion_length": 94.359375,
      "epoch": 0.6692456479690522,
      "grad_norm": 32.63760757446289,
      "kl": 0.127197265625,
      "learning_rate": 6.653771760154738e-07,
      "loss": 0.0051,
      "reward": 1.6056684255599976,
      "reward_std": 0.07967904582619667,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6056684255599976,
      "step": 1038
    },
    {
      "completion_length": 115.8125,
      "epoch": 0.6698903932946486,
      "grad_norm": 14.944701194763184,
      "kl": 0.11181640625,
      "learning_rate": 6.650548033526756e-07,
      "loss": 0.0045,
      "reward": 1.5069088339805603,
      "reward_std": 0.0734553337097168,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5069088041782379,
      "step": 1039
    },
    {
      "completion_length": 105.09375,
      "epoch": 0.670535138620245,
      "grad_norm": 8.643447875976562,
      "kl": 0.104736328125,
      "learning_rate": 6.647324306898775e-07,
      "loss": 0.0042,
      "reward": 1.4595961570739746,
      "reward_std": 0.105046346783638,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4752211719751358,
      "step": 1040
    },
    {
      "completion_length": 99.65625,
      "epoch": 0.6711798839458414,
      "grad_norm": 161.3411865234375,
      "kl": 0.119140625,
      "learning_rate": 6.644100580270793e-07,
      "loss": 0.0048,
      "reward": 1.4072444438934326,
      "reward_std": 0.12566466629505157,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4072444587945938,
      "step": 1041
    },
    {
      "completion_length": 101.953125,
      "epoch": 0.6718246292714378,
      "grad_norm": 12.601658821105957,
      "kl": 0.11767578125,
      "learning_rate": 6.640876853642811e-07,
      "loss": 0.0047,
      "reward": 1.7041231393814087,
      "reward_std": 0.08998357132077217,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7041231989860535,
      "step": 1042
    },
    {
      "completion_length": 108.03125,
      "epoch": 0.6724693745970342,
      "grad_norm": 30.03297233581543,
      "kl": 0.11767578125,
      "learning_rate": 6.637653127014829e-07,
      "loss": 0.0047,
      "reward": 1.4635764956474304,
      "reward_std": 0.1103704385459423,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4792014956474304,
      "step": 1043
    },
    {
      "completion_length": 109.25,
      "epoch": 0.6731141199226306,
      "grad_norm": 16.924476623535156,
      "kl": 0.193603515625,
      "learning_rate": 6.634429400386847e-07,
      "loss": 0.0077,
      "reward": 1.6905344724655151,
      "reward_std": 0.15296810120344162,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6905345618724823,
      "step": 1044
    },
    {
      "completion_length": 112.5625,
      "epoch": 0.6737588652482269,
      "grad_norm": 12.32383918762207,
      "kl": 0.11865234375,
      "learning_rate": 6.631205673758866e-07,
      "loss": 0.0048,
      "reward": 1.307143211364746,
      "reward_std": 0.08060317486524582,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3071431890130043,
      "step": 1045
    },
    {
      "completion_length": 116.546875,
      "epoch": 0.6744036105738234,
      "grad_norm": 18.390422821044922,
      "kl": 0.111083984375,
      "learning_rate": 6.627981947130883e-07,
      "loss": 0.0044,
      "reward": 1.497663140296936,
      "reward_std": 0.053974613547325134,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49766309559345245,
      "step": 1046
    },
    {
      "completion_length": 115.375,
      "epoch": 0.6750483558994197,
      "grad_norm": 13.646191596984863,
      "kl": 0.1611328125,
      "learning_rate": 6.624758220502901e-07,
      "loss": 0.0064,
      "reward": 1.7063047885894775,
      "reward_std": 0.16738375276327133,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7219298183917999,
      "step": 1047
    },
    {
      "completion_length": 117.125,
      "epoch": 0.6756931012250161,
      "grad_norm": 16.76261329650879,
      "kl": 0.129150390625,
      "learning_rate": 6.62153449387492e-07,
      "loss": 0.0052,
      "reward": 1.7379147410392761,
      "reward_std": 0.0969013050198555,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.737914651632309,
      "step": 1048
    },
    {
      "completion_length": 106.921875,
      "epoch": 0.6763378465506125,
      "grad_norm": 23.75704002380371,
      "kl": 0.109619140625,
      "learning_rate": 6.618310767246938e-07,
      "loss": 0.0044,
      "reward": 1.598206639289856,
      "reward_std": 0.12298348918557167,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5982065498828888,
      "step": 1049
    },
    {
      "completion_length": 122.203125,
      "epoch": 0.6769825918762089,
      "grad_norm": 12.665209770202637,
      "kl": 0.101318359375,
      "learning_rate": 6.615087040618955e-07,
      "loss": 0.0041,
      "reward": 1.7609497904777527,
      "reward_std": 0.085877925157547,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7609497904777527,
      "step": 1050
    },
    {
      "completion_length": 108.734375,
      "epoch": 0.6776273372018052,
      "grad_norm": 23.685190200805664,
      "kl": 0.112060546875,
      "learning_rate": 6.611863313990973e-07,
      "loss": 0.0045,
      "reward": 1.5682024359703064,
      "reward_std": 0.07944511994719505,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5682024210691452,
      "step": 1051
    },
    {
      "completion_length": 114.78125,
      "epoch": 0.6782720825274017,
      "grad_norm": 21.806703567504883,
      "kl": 0.107421875,
      "learning_rate": 6.608639587362991e-07,
      "loss": 0.0043,
      "reward": 1.6686413288116455,
      "reward_std": 0.07967701368033886,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6686412990093231,
      "step": 1052
    },
    {
      "completion_length": 117.140625,
      "epoch": 0.6789168278529981,
      "grad_norm": 19.198450088500977,
      "kl": 0.102783203125,
      "learning_rate": 6.60541586073501e-07,
      "loss": 0.0041,
      "reward": 1.403226375579834,
      "reward_std": 0.08735075406730175,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4032263904809952,
      "step": 1053
    },
    {
      "completion_length": 119.078125,
      "epoch": 0.6795615731785944,
      "grad_norm": 29.204330444335938,
      "kl": 0.111328125,
      "learning_rate": 6.602192134107027e-07,
      "loss": 0.0045,
      "reward": 1.7787442207336426,
      "reward_std": 0.054605141282081604,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7787441611289978,
      "step": 1054
    },
    {
      "completion_length": 112.46875,
      "epoch": 0.6802063185041909,
      "grad_norm": 11.750280380249023,
      "kl": 0.11669921875,
      "learning_rate": 6.598968407479046e-07,
      "loss": 0.0047,
      "reward": 1.4942984580993652,
      "reward_std": 0.08487318456172943,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49429842829704285,
      "step": 1055
    },
    {
      "completion_length": 118.5,
      "epoch": 0.6808510638297872,
      "grad_norm": 10.436887741088867,
      "kl": 0.109130859375,
      "learning_rate": 6.595744680851063e-07,
      "loss": 0.0044,
      "reward": 1.4903525114059448,
      "reward_std": 0.08176442608237267,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4903525114059448,
      "step": 1056
    },
    {
      "completion_length": 109.078125,
      "epoch": 0.6814958091553837,
      "grad_norm": 28.66140365600586,
      "kl": 0.1064453125,
      "learning_rate": 6.592520954223082e-07,
      "loss": 0.0043,
      "reward": 1.6072522401809692,
      "reward_std": 0.0758088231086731,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6072521805763245,
      "step": 1057
    },
    {
      "completion_length": 109.21875,
      "epoch": 0.68214055448098,
      "grad_norm": 26.009347915649414,
      "kl": 0.1083984375,
      "learning_rate": 6.5892972275951e-07,
      "loss": 0.0043,
      "reward": 1.437576413154602,
      "reward_std": 0.06020371802151203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4375763535499573,
      "step": 1058
    },
    {
      "completion_length": 123.9375,
      "epoch": 0.6827852998065764,
      "grad_norm": 44.55741500854492,
      "kl": 0.105224609375,
      "learning_rate": 6.586073500967118e-07,
      "loss": 0.0042,
      "reward": 1.5399547219276428,
      "reward_std": 0.1357093583792448,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.555579662322998,
      "step": 1059
    },
    {
      "completion_length": 118.828125,
      "epoch": 0.6834300451321728,
      "grad_norm": 83.01713562011719,
      "kl": 0.109130859375,
      "learning_rate": 6.582849774339135e-07,
      "loss": 0.0044,
      "reward": 1.545635998249054,
      "reward_std": 0.08544642850756645,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5456359833478928,
      "step": 1060
    },
    {
      "completion_length": 130.25,
      "epoch": 0.6840747904577692,
      "grad_norm": 13.197311401367188,
      "kl": 0.10693359375,
      "learning_rate": 6.579626047711155e-07,
      "loss": 0.0043,
      "reward": 1.3019261956214905,
      "reward_std": 0.06299575977027416,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3019261509180069,
      "step": 1061
    },
    {
      "completion_length": 109.578125,
      "epoch": 0.6847195357833655,
      "grad_norm": 23.34103012084961,
      "kl": 0.107177734375,
      "learning_rate": 6.576402321083172e-07,
      "loss": 0.0043,
      "reward": 1.4681254625320435,
      "reward_std": 0.10434725135564804,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4681254029273987,
      "step": 1062
    },
    {
      "completion_length": 124.90625,
      "epoch": 0.685364281108962,
      "grad_norm": 27.485233306884766,
      "kl": 0.11669921875,
      "learning_rate": 6.57317859445519e-07,
      "loss": 0.0047,
      "reward": 1.5611656308174133,
      "reward_std": 0.09634355828166008,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5611656308174133,
      "step": 1063
    },
    {
      "completion_length": 113.40625,
      "epoch": 0.6860090264345583,
      "grad_norm": 37.157684326171875,
      "kl": 0.130859375,
      "learning_rate": 6.569954867827208e-07,
      "loss": 0.0052,
      "reward": 1.490740954875946,
      "reward_std": 0.07894630171358585,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49074095487594604,
      "step": 1064
    },
    {
      "completion_length": 125.75,
      "epoch": 0.6866537717601547,
      "grad_norm": 9.309281349182129,
      "kl": 0.103759765625,
      "learning_rate": 6.566731141199227e-07,
      "loss": 0.0041,
      "reward": 1.5613402128219604,
      "reward_std": 0.10322123765945435,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5769652426242828,
      "step": 1065
    },
    {
      "completion_length": 130.375,
      "epoch": 0.6872985170857512,
      "grad_norm": 8.706680297851562,
      "kl": 0.163818359375,
      "learning_rate": 6.563507414571244e-07,
      "loss": 0.0066,
      "reward": 1.732495129108429,
      "reward_std": 0.0963818859308958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7324950993061066,
      "step": 1066
    },
    {
      "completion_length": 126.890625,
      "epoch": 0.6879432624113475,
      "grad_norm": 54.084346771240234,
      "kl": 0.125,
      "learning_rate": 6.560283687943263e-07,
      "loss": 0.005,
      "reward": 1.4755541682243347,
      "reward_std": 0.0856284536421299,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47555410861968994,
      "step": 1067
    },
    {
      "completion_length": 116.8125,
      "epoch": 0.688588007736944,
      "grad_norm": 19.93977165222168,
      "kl": 0.09814453125,
      "learning_rate": 6.55705996131528e-07,
      "loss": 0.0039,
      "reward": 1.724917709827423,
      "reward_std": 0.12488729506731033,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7405427396297455,
      "step": 1068
    },
    {
      "completion_length": 119.59375,
      "epoch": 0.6892327530625403,
      "grad_norm": 18.106996536254883,
      "kl": 0.106201171875,
      "learning_rate": 6.553836234687298e-07,
      "loss": 0.0042,
      "reward": 1.5572913885116577,
      "reward_std": 0.06482551246881485,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5572913736104965,
      "step": 1069
    },
    {
      "completion_length": 116.484375,
      "epoch": 0.6898774983881367,
      "grad_norm": 34.54378128051758,
      "kl": 0.146728515625,
      "learning_rate": 6.550612508059317e-07,
      "loss": 0.0059,
      "reward": 1.5824349522590637,
      "reward_std": 0.1754782274365425,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5980599820613861,
      "step": 1070
    },
    {
      "completion_length": 125.96875,
      "epoch": 0.690522243713733,
      "grad_norm": 14.766402244567871,
      "kl": 0.1083984375,
      "learning_rate": 6.547388781431335e-07,
      "loss": 0.0043,
      "reward": 1.4851022958755493,
      "reward_std": 0.07177170366048813,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4851022809743881,
      "step": 1071
    },
    {
      "completion_length": 108.359375,
      "epoch": 0.6911669890393295,
      "grad_norm": 17.350265502929688,
      "kl": 0.14404296875,
      "learning_rate": 6.544165054803352e-07,
      "loss": 0.0058,
      "reward": 1.5002387762069702,
      "reward_std": 0.15507403016090393,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5158638060092926,
      "step": 1072
    },
    {
      "completion_length": 111.28125,
      "epoch": 0.6918117343649258,
      "grad_norm": 26.06131935119629,
      "kl": 0.13671875,
      "learning_rate": 6.54094132817537e-07,
      "loss": 0.0054,
      "reward": 1.6340994238853455,
      "reward_std": 0.11746730655431747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6340994238853455,
      "step": 1073
    },
    {
      "completion_length": 114.84375,
      "epoch": 0.6924564796905223,
      "grad_norm": 20.31998062133789,
      "kl": 0.233642578125,
      "learning_rate": 6.537717601547389e-07,
      "loss": 0.0094,
      "reward": 1.6390128135681152,
      "reward_std": 0.08344031870365143,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6390127837657928,
      "step": 1074
    },
    {
      "completion_length": 117.546875,
      "epoch": 0.6931012250161186,
      "grad_norm": 12.814098358154297,
      "kl": 0.120361328125,
      "learning_rate": 6.534493874919407e-07,
      "loss": 0.0048,
      "reward": 1.5889630317687988,
      "reward_std": 0.08722979947924614,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5889630317687988,
      "step": 1075
    },
    {
      "completion_length": 108.6875,
      "epoch": 0.693745970341715,
      "grad_norm": 34.06943893432617,
      "kl": 0.118896484375,
      "learning_rate": 6.531270148291424e-07,
      "loss": 0.0048,
      "reward": 1.549480378627777,
      "reward_std": 0.08187295310199261,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5494804084300995,
      "step": 1076
    },
    {
      "completion_length": 115.671875,
      "epoch": 0.6943907156673114,
      "grad_norm": 17.84603500366211,
      "kl": 0.144775390625,
      "learning_rate": 6.528046421663443e-07,
      "loss": 0.0058,
      "reward": 1.702250897884369,
      "reward_std": 0.09670410864055157,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7022508978843689,
      "step": 1077
    },
    {
      "completion_length": 118.46875,
      "epoch": 0.6950354609929078,
      "grad_norm": 15.43232250213623,
      "kl": 0.12158203125,
      "learning_rate": 6.524822695035461e-07,
      "loss": 0.0049,
      "reward": 1.7633506059646606,
      "reward_std": 0.08454957231879234,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7633506059646606,
      "step": 1078
    },
    {
      "completion_length": 120.390625,
      "epoch": 0.6956802063185042,
      "grad_norm": 19.23419761657715,
      "kl": 0.1318359375,
      "learning_rate": 6.521598968407479e-07,
      "loss": 0.0053,
      "reward": 1.5779541730880737,
      "reward_std": 0.08321835845708847,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5779542624950409,
      "step": 1079
    },
    {
      "completion_length": 111.40625,
      "epoch": 0.6963249516441006,
      "grad_norm": 44.08869934082031,
      "kl": 0.134521484375,
      "learning_rate": 6.518375241779497e-07,
      "loss": 0.0054,
      "reward": 1.6062058210372925,
      "reward_std": 0.09365630894899368,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6062057018280029,
      "step": 1080
    },
    {
      "completion_length": 125.3125,
      "epoch": 0.696969696969697,
      "grad_norm": 9.416675567626953,
      "kl": 0.118408203125,
      "learning_rate": 6.515151515151515e-07,
      "loss": 0.0047,
      "reward": 1.6733117699623108,
      "reward_std": 0.09015357866883278,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.673311859369278,
      "step": 1081
    },
    {
      "completion_length": 111.15625,
      "epoch": 0.6976144422952933,
      "grad_norm": 5.396347522735596,
      "kl": 0.110595703125,
      "learning_rate": 6.511927788523532e-07,
      "loss": 0.0044,
      "reward": 1.5660415291786194,
      "reward_std": 0.13443563878536224,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5972915291786194,
      "step": 1082
    },
    {
      "completion_length": 124.765625,
      "epoch": 0.6982591876208898,
      "grad_norm": 19.220321655273438,
      "kl": 0.12890625,
      "learning_rate": 6.508704061895552e-07,
      "loss": 0.0051,
      "reward": 1.4645467400550842,
      "reward_std": 0.09206113591790199,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4645467698574066,
      "step": 1083
    },
    {
      "completion_length": 119.125,
      "epoch": 0.6989039329464861,
      "grad_norm": 9.741296768188477,
      "kl": 0.113525390625,
      "learning_rate": 6.505480335267569e-07,
      "loss": 0.0045,
      "reward": 1.7223733067512512,
      "reward_std": 0.14363746345043182,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7223733961582184,
      "step": 1084
    },
    {
      "completion_length": 115.359375,
      "epoch": 0.6995486782720826,
      "grad_norm": 10.412392616271973,
      "kl": 0.11474609375,
      "learning_rate": 6.502256608639587e-07,
      "loss": 0.0046,
      "reward": 1.634882390499115,
      "reward_std": 0.03937092795968056,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6348824203014374,
      "step": 1085
    },
    {
      "completion_length": 111.078125,
      "epoch": 0.7001934235976789,
      "grad_norm": 52.82218551635742,
      "kl": 0.17236328125,
      "learning_rate": 6.499032882011605e-07,
      "loss": 0.0069,
      "reward": 1.6078645586967468,
      "reward_std": 0.14649196714162827,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.607864499092102,
      "step": 1086
    },
    {
      "completion_length": 110.625,
      "epoch": 0.7008381689232753,
      "grad_norm": 10.968231201171875,
      "kl": 0.11279296875,
      "learning_rate": 6.495809155383624e-07,
      "loss": 0.0045,
      "reward": 1.4408124089241028,
      "reward_std": 0.05924747884273529,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4408123791217804,
      "step": 1087
    },
    {
      "completion_length": 106.765625,
      "epoch": 0.7014829142488717,
      "grad_norm": 123.17638397216797,
      "kl": 0.127685546875,
      "learning_rate": 6.492585428755641e-07,
      "loss": 0.0051,
      "reward": 1.5098558068275452,
      "reward_std": 0.08591609820723534,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.509855791926384,
      "step": 1088
    },
    {
      "completion_length": 101.53125,
      "epoch": 0.7021276595744681,
      "grad_norm": 17.40302085876465,
      "kl": 0.13330078125,
      "learning_rate": 6.48936170212766e-07,
      "loss": 0.0053,
      "reward": 1.7117857933044434,
      "reward_std": 0.10072560980916023,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7117857933044434,
      "step": 1089
    },
    {
      "completion_length": 111.296875,
      "epoch": 0.7027724049000644,
      "grad_norm": 13.590946197509766,
      "kl": 0.10986328125,
      "learning_rate": 6.486137975499677e-07,
      "loss": 0.0044,
      "reward": 1.6580272912979126,
      "reward_std": 0.06897935643792152,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6580273658037186,
      "step": 1090
    },
    {
      "completion_length": 96.625,
      "epoch": 0.7034171502256609,
      "grad_norm": 30.83803367614746,
      "kl": 0.100830078125,
      "learning_rate": 6.482914248871696e-07,
      "loss": 0.004,
      "reward": 1.6532940864562988,
      "reward_std": 0.15903402864933014,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6689190864562988,
      "step": 1091
    },
    {
      "completion_length": 105.78125,
      "epoch": 0.7040618955512572,
      "grad_norm": 25.37020492553711,
      "kl": 0.117431640625,
      "learning_rate": 6.479690522243714e-07,
      "loss": 0.0047,
      "reward": 1.497763454914093,
      "reward_std": 0.09194651618599892,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49776342511177063,
      "step": 1092
    },
    {
      "completion_length": 114.203125,
      "epoch": 0.7047066408768536,
      "grad_norm": 13.3228120803833,
      "kl": 0.1279296875,
      "learning_rate": 6.476466795615732e-07,
      "loss": 0.0051,
      "reward": 1.725711703300476,
      "reward_std": 0.1199619323015213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7257117033004761,
      "step": 1093
    },
    {
      "completion_length": 121.0625,
      "epoch": 0.7053513862024501,
      "grad_norm": 12.090010643005371,
      "kl": 0.1220703125,
      "learning_rate": 6.473243068987749e-07,
      "loss": 0.0049,
      "reward": 1.4209555387496948,
      "reward_std": 0.15032630413770676,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.43658050894737244,
      "step": 1094
    },
    {
      "completion_length": 118.359375,
      "epoch": 0.7059961315280464,
      "grad_norm": 21.097448348999023,
      "kl": 0.115966796875,
      "learning_rate": 6.470019342359767e-07,
      "loss": 0.0046,
      "reward": 1.6207126379013062,
      "reward_std": 0.11108803562819958,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6363376677036285,
      "step": 1095
    },
    {
      "completion_length": 108.453125,
      "epoch": 0.7066408768536429,
      "grad_norm": 11.777490615844727,
      "kl": 0.121337890625,
      "learning_rate": 6.466795615731786e-07,
      "loss": 0.0049,
      "reward": 1.5517085194587708,
      "reward_std": 0.11031496897339821,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5517085194587708,
      "step": 1096
    },
    {
      "completion_length": 109.59375,
      "epoch": 0.7072856221792392,
      "grad_norm": 7.2100419998168945,
      "kl": 0.11328125,
      "learning_rate": 6.463571889103804e-07,
      "loss": 0.0045,
      "reward": 1.5937931537628174,
      "reward_std": 0.06582435220479965,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5937931537628174,
      "step": 1097
    },
    {
      "completion_length": 105.90625,
      "epoch": 0.7079303675048356,
      "grad_norm": 65.8556137084961,
      "kl": 0.1201171875,
      "learning_rate": 6.460348162475821e-07,
      "loss": 0.0048,
      "reward": 1.4370640516281128,
      "reward_std": 0.16717902943491936,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.4683140963315964,
      "step": 1098
    },
    {
      "completion_length": 101.015625,
      "epoch": 0.708575112830432,
      "grad_norm": 17.487159729003906,
      "kl": 0.113037109375,
      "learning_rate": 6.45712443584784e-07,
      "loss": 0.0045,
      "reward": 1.500916838645935,
      "reward_std": 0.053837522864341736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5009167790412903,
      "step": 1099
    },
    {
      "completion_length": 115.875,
      "epoch": 0.7092198581560284,
      "grad_norm": 11.76062297821045,
      "kl": 0.11376953125,
      "learning_rate": 6.453900709219858e-07,
      "loss": 0.0045,
      "reward": 1.4913934469223022,
      "reward_std": 0.13147876039147377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49139344692230225,
      "step": 1100
    },
    {
      "completion_length": 109.390625,
      "epoch": 0.7098646034816247,
      "grad_norm": 8.682186126708984,
      "kl": 0.117919921875,
      "learning_rate": 6.450676982591876e-07,
      "loss": 0.0047,
      "reward": 1.6740834712982178,
      "reward_std": 0.14168205112218857,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6897085309028625,
      "step": 1101
    },
    {
      "completion_length": 97.09375,
      "epoch": 0.7105093488072212,
      "grad_norm": 15.599982261657715,
      "kl": 0.131103515625,
      "learning_rate": 6.447453255963894e-07,
      "loss": 0.0052,
      "reward": 1.4697246551513672,
      "reward_std": 0.05679570883512497,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4697246551513672,
      "step": 1102
    },
    {
      "completion_length": 103.953125,
      "epoch": 0.7111540941328175,
      "grad_norm": 12.54194450378418,
      "kl": 0.111083984375,
      "learning_rate": 6.444229529335912e-07,
      "loss": 0.0044,
      "reward": 1.5123698711395264,
      "reward_std": 0.13531113602221012,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5123699456453323,
      "step": 1103
    },
    {
      "completion_length": 97.921875,
      "epoch": 0.7117988394584139,
      "grad_norm": 18.03816795349121,
      "kl": 0.1318359375,
      "learning_rate": 6.44100580270793e-07,
      "loss": 0.0053,
      "reward": 1.6060336232185364,
      "reward_std": 0.08316031470894814,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6060336530208588,
      "step": 1104
    },
    {
      "completion_length": 105.875,
      "epoch": 0.7124435847840103,
      "grad_norm": 15.154292106628418,
      "kl": 0.11865234375,
      "learning_rate": 6.437782076079949e-07,
      "loss": 0.0048,
      "reward": 1.5886054635047913,
      "reward_std": 0.1678875982761383,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6042304337024689,
      "step": 1105
    },
    {
      "completion_length": 90.84375,
      "epoch": 0.7130883301096067,
      "grad_norm": 9.316852569580078,
      "kl": 0.126953125,
      "learning_rate": 6.434558349451966e-07,
      "loss": 0.0051,
      "reward": 1.2697542905807495,
      "reward_std": 0.0764620453119278,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.2697542533278465,
      "step": 1106
    },
    {
      "completion_length": 103.5625,
      "epoch": 0.7137330754352031,
      "grad_norm": 21.145282745361328,
      "kl": 0.11181640625,
      "learning_rate": 6.431334622823984e-07,
      "loss": 0.0045,
      "reward": 1.679430365562439,
      "reward_std": 0.06132471561431885,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6794304251670837,
      "step": 1107
    },
    {
      "completion_length": 109.71875,
      "epoch": 0.7143778207607995,
      "grad_norm": 13.92724323272705,
      "kl": 0.12255859375,
      "learning_rate": 6.428110896196002e-07,
      "loss": 0.0049,
      "reward": 1.6119479537010193,
      "reward_std": 0.12308447062969208,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6119480133056641,
      "step": 1108
    },
    {
      "completion_length": 105.0,
      "epoch": 0.7150225660863959,
      "grad_norm": 170.48899841308594,
      "kl": 0.122802734375,
      "learning_rate": 6.424887169568021e-07,
      "loss": 0.0049,
      "reward": 1.5171204805374146,
      "reward_std": 0.09489506855607033,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5171204507350922,
      "step": 1109
    },
    {
      "completion_length": 107.46875,
      "epoch": 0.7156673114119922,
      "grad_norm": 40.21750259399414,
      "kl": 0.11669921875,
      "learning_rate": 6.421663442940038e-07,
      "loss": 0.0047,
      "reward": 1.6050112843513489,
      "reward_std": 0.08443249017000198,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6050112396478653,
      "step": 1110
    },
    {
      "completion_length": 103.4375,
      "epoch": 0.7163120567375887,
      "grad_norm": 7.109347820281982,
      "kl": 0.114013671875,
      "learning_rate": 6.418439716312057e-07,
      "loss": 0.0046,
      "reward": 1.4758098721504211,
      "reward_std": 0.05706555396318436,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4758097976446152,
      "step": 1111
    },
    {
      "completion_length": 114.15625,
      "epoch": 0.716956802063185,
      "grad_norm": 8.961230278015137,
      "kl": 0.132080078125,
      "learning_rate": 6.415215989684074e-07,
      "loss": 0.0053,
      "reward": 1.5424306988716125,
      "reward_std": 0.10397448390722275,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5424306988716125,
      "step": 1112
    },
    {
      "completion_length": 107.0,
      "epoch": 0.7176015473887815,
      "grad_norm": 7.983897686004639,
      "kl": 0.1279296875,
      "learning_rate": 6.411992263056093e-07,
      "loss": 0.0051,
      "reward": 1.432960569858551,
      "reward_std": 0.05943846330046654,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.432960569858551,
      "step": 1113
    },
    {
      "completion_length": 105.1875,
      "epoch": 0.7182462927143778,
      "grad_norm": 9.793622016906738,
      "kl": 0.12646484375,
      "learning_rate": 6.40876853642811e-07,
      "loss": 0.0051,
      "reward": 1.5297775864601135,
      "reward_std": 0.09795258566737175,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5297775864601135,
      "step": 1114
    },
    {
      "completion_length": 100.65625,
      "epoch": 0.7188910380399742,
      "grad_norm": 21.03094482421875,
      "kl": 0.11083984375,
      "learning_rate": 6.405544809800129e-07,
      "loss": 0.0044,
      "reward": 1.5464897155761719,
      "reward_std": 0.12115862965583801,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5464896857738495,
      "step": 1115
    },
    {
      "completion_length": 106.25,
      "epoch": 0.7195357833655706,
      "grad_norm": 23.36917495727539,
      "kl": 0.123046875,
      "learning_rate": 6.402321083172146e-07,
      "loss": 0.0049,
      "reward": 1.4555120468139648,
      "reward_std": 0.09448462724685669,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.47113706171512604,
      "step": 1116
    },
    {
      "completion_length": 98.09375,
      "epoch": 0.720180528691167,
      "grad_norm": 10.004223823547363,
      "kl": 0.122314453125,
      "learning_rate": 6.399097356544166e-07,
      "loss": 0.0049,
      "reward": 1.3829104900360107,
      "reward_std": 0.06856898404657841,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.38291044533252716,
      "step": 1117
    },
    {
      "completion_length": 95.671875,
      "epoch": 0.7208252740167633,
      "grad_norm": 137.10011291503906,
      "kl": 0.13818359375,
      "learning_rate": 6.395873629916183e-07,
      "loss": 0.0055,
      "reward": 1.5769847631454468,
      "reward_std": 0.06603378802537918,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.576984703540802,
      "step": 1118
    },
    {
      "completion_length": 109.03125,
      "epoch": 0.7214700193423598,
      "grad_norm": 17.249679565429688,
      "kl": 0.1142578125,
      "learning_rate": 6.392649903288201e-07,
      "loss": 0.0046,
      "reward": 1.624714195728302,
      "reward_std": 0.1402108147740364,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6247142255306244,
      "step": 1119
    },
    {
      "completion_length": 106.84375,
      "epoch": 0.7221147646679562,
      "grad_norm": 9.263527870178223,
      "kl": 0.15380859375,
      "learning_rate": 6.389426176660218e-07,
      "loss": 0.0062,
      "reward": 1.6390187740325928,
      "reward_std": 0.12167927622795105,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6390187740325928,
      "step": 1120
    },
    {
      "completion_length": 114.359375,
      "epoch": 0.7227595099935525,
      "grad_norm": 26.278076171875,
      "kl": 0.1640625,
      "learning_rate": 6.386202450032237e-07,
      "loss": 0.0066,
      "reward": 1.6548975110054016,
      "reward_std": 0.16698498651385307,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6705225110054016,
      "step": 1121
    },
    {
      "completion_length": 106.515625,
      "epoch": 0.723404255319149,
      "grad_norm": 41.87917709350586,
      "kl": 0.1552734375,
      "learning_rate": 6.382978723404255e-07,
      "loss": 0.0062,
      "reward": 1.6121267080307007,
      "reward_std": 0.10054780542850494,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6121267527341843,
      "step": 1122
    },
    {
      "completion_length": 95.09375,
      "epoch": 0.7240490006447453,
      "grad_norm": 17.717201232910156,
      "kl": 0.1796875,
      "learning_rate": 6.379754996776273e-07,
      "loss": 0.0072,
      "reward": 1.5679856538772583,
      "reward_std": 0.08641625568270683,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5679856687784195,
      "step": 1123
    },
    {
      "completion_length": 103.8125,
      "epoch": 0.7246937459703418,
      "grad_norm": 10.81406021118164,
      "kl": 0.1337890625,
      "learning_rate": 6.376531270148291e-07,
      "loss": 0.0054,
      "reward": 1.4295477271080017,
      "reward_std": 0.08662409894168377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4295477420091629,
      "step": 1124
    },
    {
      "completion_length": 119.09375,
      "epoch": 0.7253384912959381,
      "grad_norm": 8.933006286621094,
      "kl": 0.11181640625,
      "learning_rate": 6.373307543520309e-07,
      "loss": 0.0045,
      "reward": 1.4276261925697327,
      "reward_std": 0.10313639417290688,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42762622237205505,
      "step": 1125
    },
    {
      "completion_length": 109.359375,
      "epoch": 0.7259832366215345,
      "grad_norm": 31.597089767456055,
      "kl": 0.1337890625,
      "learning_rate": 6.370083816892327e-07,
      "loss": 0.0053,
      "reward": 1.6690808534622192,
      "reward_std": 0.0815388485789299,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6690808236598969,
      "step": 1126
    },
    {
      "completion_length": 107.15625,
      "epoch": 0.7266279819471309,
      "grad_norm": 18.734527587890625,
      "kl": 0.1337890625,
      "learning_rate": 6.366860090264346e-07,
      "loss": 0.0054,
      "reward": 1.4915392994880676,
      "reward_std": 0.1256212517619133,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49153923988342285,
      "step": 1127
    },
    {
      "completion_length": 103.609375,
      "epoch": 0.7272727272727273,
      "grad_norm": 10.532417297363281,
      "kl": 0.138671875,
      "learning_rate": 6.363636363636363e-07,
      "loss": 0.0055,
      "reward": 1.6402064561843872,
      "reward_std": 0.08831041865050793,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6402064859867096,
      "step": 1128
    },
    {
      "completion_length": 116.015625,
      "epoch": 0.7279174725983236,
      "grad_norm": 70.3717041015625,
      "kl": 0.1162109375,
      "learning_rate": 6.360412637008381e-07,
      "loss": 0.0046,
      "reward": 1.5526406168937683,
      "reward_std": 0.09058434516191483,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5526406168937683,
      "step": 1129
    },
    {
      "completion_length": 99.296875,
      "epoch": 0.7285622179239201,
      "grad_norm": 22.534345626831055,
      "kl": 0.1533203125,
      "learning_rate": 6.3571889103804e-07,
      "loss": 0.0061,
      "reward": 1.6604415774345398,
      "reward_std": 0.11532147601246834,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6604415774345398,
      "step": 1130
    },
    {
      "completion_length": 104.046875,
      "epoch": 0.7292069632495164,
      "grad_norm": 11.422441482543945,
      "kl": 0.11767578125,
      "learning_rate": 6.353965183752418e-07,
      "loss": 0.0047,
      "reward": 1.6915850639343262,
      "reward_std": 0.11195739358663559,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6915850639343262,
      "step": 1131
    },
    {
      "completion_length": 112.078125,
      "epoch": 0.7298517085751128,
      "grad_norm": 19.152061462402344,
      "kl": 0.13525390625,
      "learning_rate": 6.350741457124435e-07,
      "loss": 0.0054,
      "reward": 1.5175195932388306,
      "reward_std": 0.060434307903051376,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5175195336341858,
      "step": 1132
    },
    {
      "completion_length": 111.328125,
      "epoch": 0.7304964539007093,
      "grad_norm": 14.260641098022461,
      "kl": 0.16552734375,
      "learning_rate": 6.347517730496454e-07,
      "loss": 0.0066,
      "reward": 1.5333157181739807,
      "reward_std": 0.10276435315608978,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5333157181739807,
      "step": 1133
    },
    {
      "completion_length": 100.328125,
      "epoch": 0.7311411992263056,
      "grad_norm": 24.769067764282227,
      "kl": 0.12939453125,
      "learning_rate": 6.344294003868471e-07,
      "loss": 0.0052,
      "reward": 1.8205824494361877,
      "reward_std": 0.0905773900449276,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.8205824196338654,
      "step": 1134
    },
    {
      "completion_length": 113.640625,
      "epoch": 0.731785944551902,
      "grad_norm": 25.011865615844727,
      "kl": 0.13134765625,
      "learning_rate": 6.34107027724049e-07,
      "loss": 0.0053,
      "reward": 1.6984012126922607,
      "reward_std": 0.06417006999254227,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6984012126922607,
      "step": 1135
    },
    {
      "completion_length": 106.515625,
      "epoch": 0.7324306898774984,
      "grad_norm": 21.090362548828125,
      "kl": 0.12353515625,
      "learning_rate": 6.337846550612508e-07,
      "loss": 0.0049,
      "reward": 1.5312876105308533,
      "reward_std": 0.07693489640951157,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5312876403331757,
      "step": 1136
    },
    {
      "completion_length": 121.703125,
      "epoch": 0.7330754352030948,
      "grad_norm": 11.780389785766602,
      "kl": 0.12646484375,
      "learning_rate": 6.334622823984526e-07,
      "loss": 0.005,
      "reward": 1.6064636707305908,
      "reward_std": 0.10624757781624794,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6064636707305908,
      "step": 1137
    },
    {
      "completion_length": 113.671875,
      "epoch": 0.7337201805286911,
      "grad_norm": 24.819656372070312,
      "kl": 0.113525390625,
      "learning_rate": 6.331399097356543e-07,
      "loss": 0.0045,
      "reward": 1.6082735657691956,
      "reward_std": 0.10192062892019749,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6238984763622284,
      "step": 1138
    },
    {
      "completion_length": 98.25,
      "epoch": 0.7343649258542876,
      "grad_norm": 11.147581100463867,
      "kl": 0.109130859375,
      "learning_rate": 6.328175370728563e-07,
      "loss": 0.0044,
      "reward": 1.5456410646438599,
      "reward_std": 0.07163537293672562,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5456410944461823,
      "step": 1139
    },
    {
      "completion_length": 108.734375,
      "epoch": 0.7350096711798839,
      "grad_norm": 16.347904205322266,
      "kl": 0.12158203125,
      "learning_rate": 6.32495164410058e-07,
      "loss": 0.0049,
      "reward": 1.7426636219024658,
      "reward_std": 0.08242497406899929,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7426636517047882,
      "step": 1140
    },
    {
      "completion_length": 115.0,
      "epoch": 0.7356544165054804,
      "grad_norm": 24.418312072753906,
      "kl": 0.1181640625,
      "learning_rate": 6.321727917472598e-07,
      "loss": 0.0047,
      "reward": 1.4525124430656433,
      "reward_std": 0.11574630439281464,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4525124728679657,
      "step": 1141
    },
    {
      "completion_length": 102.71875,
      "epoch": 0.7362991618310767,
      "grad_norm": 23.146039962768555,
      "kl": 0.12548828125,
      "learning_rate": 6.318504190844615e-07,
      "loss": 0.005,
      "reward": 1.4995726943016052,
      "reward_std": 0.09345082193613052,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4995726943016052,
      "step": 1142
    },
    {
      "completion_length": 99.359375,
      "epoch": 0.7369439071566731,
      "grad_norm": 11.854866981506348,
      "kl": 0.1611328125,
      "learning_rate": 6.315280464216635e-07,
      "loss": 0.0064,
      "reward": 1.5400469899177551,
      "reward_std": 0.152745820581913,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5556719899177551,
      "step": 1143
    },
    {
      "completion_length": 110.265625,
      "epoch": 0.7375886524822695,
      "grad_norm": 23.145206451416016,
      "kl": 0.12939453125,
      "learning_rate": 6.312056737588652e-07,
      "loss": 0.0052,
      "reward": 1.7510847449302673,
      "reward_std": 0.0897570513188839,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7510847449302673,
      "step": 1144
    },
    {
      "completion_length": 115.1875,
      "epoch": 0.7382333978078659,
      "grad_norm": 7.87054967880249,
      "kl": 0.133056640625,
      "learning_rate": 6.30883301096067e-07,
      "loss": 0.0053,
      "reward": 1.4778541326522827,
      "reward_std": 0.09418310970067978,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4778541922569275,
      "step": 1145
    },
    {
      "completion_length": 109.515625,
      "epoch": 0.7388781431334622,
      "grad_norm": 73.98107147216797,
      "kl": 0.7998046875,
      "learning_rate": 6.305609284332688e-07,
      "loss": 0.032,
      "reward": 1.671739637851715,
      "reward_std": 0.08632522821426392,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6717396378517151,
      "step": 1146
    },
    {
      "completion_length": 122.53125,
      "epoch": 0.7395228884590587,
      "grad_norm": 31.42171287536621,
      "kl": 0.128173828125,
      "learning_rate": 6.302385557704706e-07,
      "loss": 0.0051,
      "reward": 1.554942011833191,
      "reward_std": 0.14347686991095543,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5549419820308685,
      "step": 1147
    },
    {
      "completion_length": 109.09375,
      "epoch": 0.7401676337846551,
      "grad_norm": 15.380295753479004,
      "kl": 0.111083984375,
      "learning_rate": 6.299161831076724e-07,
      "loss": 0.0045,
      "reward": 1.4521582126617432,
      "reward_std": 0.05634276382625103,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45215825736522675,
      "step": 1148
    },
    {
      "completion_length": 108.84375,
      "epoch": 0.7408123791102514,
      "grad_norm": 5.173756122589111,
      "kl": 0.1240234375,
      "learning_rate": 6.295938104448743e-07,
      "loss": 0.005,
      "reward": 1.5091955661773682,
      "reward_std": 0.08003686740994453,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5091955214738846,
      "step": 1149
    },
    {
      "completion_length": 110.515625,
      "epoch": 0.7414571244358479,
      "grad_norm": 13.267820358276367,
      "kl": 0.1337890625,
      "learning_rate": 6.29271437782076e-07,
      "loss": 0.0053,
      "reward": 1.4815285801887512,
      "reward_std": 0.06610534526407719,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4815286546945572,
      "step": 1150
    },
    {
      "completion_length": 103.546875,
      "epoch": 0.7421018697614442,
      "grad_norm": 7.666851043701172,
      "kl": 0.12548828125,
      "learning_rate": 6.289490651192778e-07,
      "loss": 0.005,
      "reward": 1.4223946332931519,
      "reward_std": 0.09108829498291016,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42239460349082947,
      "step": 1151
    },
    {
      "completion_length": 112.078125,
      "epoch": 0.7427466150870407,
      "grad_norm": 46.41838836669922,
      "kl": 0.117431640625,
      "learning_rate": 6.286266924564797e-07,
      "loss": 0.0047,
      "reward": 1.3338908553123474,
      "reward_std": 0.11570664495229721,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3338908553123474,
      "step": 1152
    },
    {
      "completion_length": 105.8125,
      "epoch": 0.743391360412637,
      "grad_norm": 39.60142517089844,
      "kl": 0.121826171875,
      "learning_rate": 6.283043197936815e-07,
      "loss": 0.0049,
      "reward": 1.4156802892684937,
      "reward_std": 0.05137963965535164,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41568028926849365,
      "step": 1153
    },
    {
      "completion_length": 130.8125,
      "epoch": 0.7440361057382334,
      "grad_norm": 8.230225563049316,
      "kl": 0.111328125,
      "learning_rate": 6.279819471308832e-07,
      "loss": 0.0045,
      "reward": 1.5346974730491638,
      "reward_std": 0.13676878809928894,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5346974283456802,
      "step": 1154
    },
    {
      "completion_length": 102.15625,
      "epoch": 0.7446808510638298,
      "grad_norm": 16.30229377746582,
      "kl": 0.1484375,
      "learning_rate": 6.276595744680851e-07,
      "loss": 0.0059,
      "reward": 1.6616498827934265,
      "reward_std": 0.09401846304535866,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6616499423980713,
      "step": 1155
    },
    {
      "completion_length": 124.65625,
      "epoch": 0.7453255963894262,
      "grad_norm": 9.194766998291016,
      "kl": 0.1552734375,
      "learning_rate": 6.273372018052869e-07,
      "loss": 0.0062,
      "reward": 1.4719805121421814,
      "reward_std": 0.09085122868418694,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4876054525375366,
      "step": 1156
    },
    {
      "completion_length": 112.515625,
      "epoch": 0.7459703417150225,
      "grad_norm": 34.08546829223633,
      "kl": 0.1337890625,
      "learning_rate": 6.270148291424887e-07,
      "loss": 0.0053,
      "reward": 1.4912188053131104,
      "reward_std": 0.15397004038095474,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4912187159061432,
      "step": 1157
    },
    {
      "completion_length": 122.4375,
      "epoch": 0.746615087040619,
      "grad_norm": 7.057889461517334,
      "kl": 0.13232421875,
      "learning_rate": 6.266924564796905e-07,
      "loss": 0.0053,
      "reward": 1.6029225587844849,
      "reward_std": 0.06702636182308197,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6029225587844849,
      "step": 1158
    },
    {
      "completion_length": 108.75,
      "epoch": 0.7472598323662153,
      "grad_norm": 20.546005249023438,
      "kl": 0.128662109375,
      "learning_rate": 6.263700838168923e-07,
      "loss": 0.0052,
      "reward": 1.7207480072975159,
      "reward_std": 0.09248170256614685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7207480072975159,
      "step": 1159
    },
    {
      "completion_length": 121.984375,
      "epoch": 0.7479045776918117,
      "grad_norm": 10.88604736328125,
      "kl": 0.13818359375,
      "learning_rate": 6.26047711154094e-07,
      "loss": 0.0055,
      "reward": 1.6104546785354614,
      "reward_std": 0.14067476242780685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6104547083377838,
      "step": 1160
    },
    {
      "completion_length": 112.3125,
      "epoch": 0.7485493230174082,
      "grad_norm": 17.844881057739258,
      "kl": 0.1328125,
      "learning_rate": 6.25725338491296e-07,
      "loss": 0.0053,
      "reward": 1.4679070711135864,
      "reward_std": 0.0854264535009861,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46790711581707,
      "step": 1161
    },
    {
      "completion_length": 109.359375,
      "epoch": 0.7491940683430045,
      "grad_norm": 11.832077026367188,
      "kl": 0.1396484375,
      "learning_rate": 6.254029658284977e-07,
      "loss": 0.0056,
      "reward": 1.5666400790214539,
      "reward_std": 0.07326122745871544,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5666401386260986,
      "step": 1162
    },
    {
      "completion_length": 109.5,
      "epoch": 0.749838813668601,
      "grad_norm": 12.351283073425293,
      "kl": 0.1318359375,
      "learning_rate": 6.250805931656995e-07,
      "loss": 0.0053,
      "reward": 1.4992797374725342,
      "reward_std": 0.08922664262354374,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4992796927690506,
      "step": 1163
    },
    {
      "completion_length": 108.328125,
      "epoch": 0.7504835589941973,
      "grad_norm": 14.127129554748535,
      "kl": 0.128662109375,
      "learning_rate": 6.247582205029012e-07,
      "loss": 0.0051,
      "reward": 1.725237250328064,
      "reward_std": 0.05798189900815487,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.725237250328064,
      "step": 1164
    },
    {
      "completion_length": 109.875,
      "epoch": 0.7511283043197937,
      "grad_norm": 10.182910919189453,
      "kl": 0.16796875,
      "learning_rate": 6.244358478401032e-07,
      "loss": 0.0067,
      "reward": 1.5962581038475037,
      "reward_std": 0.11094292625784874,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.596258133649826,
      "step": 1165
    },
    {
      "completion_length": 114.96875,
      "epoch": 0.75177304964539,
      "grad_norm": 33.12942886352539,
      "kl": 0.1328125,
      "learning_rate": 6.241134751773049e-07,
      "loss": 0.0053,
      "reward": 1.474349319934845,
      "reward_std": 0.13836726546287537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47434939444065094,
      "step": 1166
    },
    {
      "completion_length": 116.828125,
      "epoch": 0.7524177949709865,
      "grad_norm": 20.074556350708008,
      "kl": 0.12646484375,
      "learning_rate": 6.237911025145067e-07,
      "loss": 0.0051,
      "reward": 1.6606274843215942,
      "reward_std": 0.09943772666156292,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.660627543926239,
      "step": 1167
    },
    {
      "completion_length": 117.546875,
      "epoch": 0.7530625402965828,
      "grad_norm": 9.653539657592773,
      "kl": 0.1259765625,
      "learning_rate": 6.234687298517085e-07,
      "loss": 0.0051,
      "reward": 1.4474668502807617,
      "reward_std": 0.1122102215886116,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4630918353796005,
      "step": 1168
    },
    {
      "completion_length": 106.90625,
      "epoch": 0.7537072856221793,
      "grad_norm": 20.613239288330078,
      "kl": 0.12744140625,
      "learning_rate": 6.231463571889104e-07,
      "loss": 0.0051,
      "reward": 1.421180248260498,
      "reward_std": 0.12799222767353058,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42118029296398163,
      "step": 1169
    },
    {
      "completion_length": 116.25,
      "epoch": 0.7543520309477756,
      "grad_norm": 10.835249900817871,
      "kl": 0.110107421875,
      "learning_rate": 6.228239845261121e-07,
      "loss": 0.0044,
      "reward": 1.6783432960510254,
      "reward_std": 0.07521459832787514,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6783432960510254,
      "step": 1170
    },
    {
      "completion_length": 113.671875,
      "epoch": 0.754996776273372,
      "grad_norm": 36.19251251220703,
      "kl": 0.146484375,
      "learning_rate": 6.22501611863314e-07,
      "loss": 0.0059,
      "reward": 1.4784401059150696,
      "reward_std": 0.13166068121790886,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4940650761127472,
      "step": 1171
    },
    {
      "completion_length": 103.25,
      "epoch": 0.7556415215989684,
      "grad_norm": 13.30659008026123,
      "kl": 0.12744140625,
      "learning_rate": 6.221792392005157e-07,
      "loss": 0.0051,
      "reward": 1.6257249116897583,
      "reward_std": 0.15445075929164886,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6413498520851135,
      "step": 1172
    },
    {
      "completion_length": 123.484375,
      "epoch": 0.7562862669245648,
      "grad_norm": 11.93229866027832,
      "kl": 0.193359375,
      "learning_rate": 6.218568665377176e-07,
      "loss": 0.0077,
      "reward": 1.6381423473358154,
      "reward_std": 0.09956448897719383,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6381423473358154,
      "step": 1173
    },
    {
      "completion_length": 111.078125,
      "epoch": 0.7569310122501612,
      "grad_norm": 26.301389694213867,
      "kl": 0.141845703125,
      "learning_rate": 6.215344938749194e-07,
      "loss": 0.0057,
      "reward": 1.527126967906952,
      "reward_std": 0.11088033393025398,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5271269381046295,
      "step": 1174
    },
    {
      "completion_length": 107.875,
      "epoch": 0.7575757575757576,
      "grad_norm": 20.331317901611328,
      "kl": 0.12841796875,
      "learning_rate": 6.212121212121212e-07,
      "loss": 0.0051,
      "reward": 1.5884113907814026,
      "reward_std": 0.12505877017974854,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5884113907814026,
      "step": 1175
    },
    {
      "completion_length": 110.65625,
      "epoch": 0.758220502901354,
      "grad_norm": 16.177221298217773,
      "kl": 0.1337890625,
      "learning_rate": 6.208897485493229e-07,
      "loss": 0.0053,
      "reward": 1.5988757014274597,
      "reward_std": 0.1253979578614235,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5988757908344269,
      "step": 1176
    },
    {
      "completion_length": 108.15625,
      "epoch": 0.7588652482269503,
      "grad_norm": 100.30497741699219,
      "kl": 0.12841796875,
      "learning_rate": 6.205673758865248e-07,
      "loss": 0.0051,
      "reward": 1.4070366621017456,
      "reward_std": 0.11220012977719307,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4070366621017456,
      "step": 1177
    },
    {
      "completion_length": 107.125,
      "epoch": 0.7595099935525468,
      "grad_norm": 9.922935485839844,
      "kl": 0.135498046875,
      "learning_rate": 6.202450032237266e-07,
      "loss": 0.0054,
      "reward": 1.6834790706634521,
      "reward_std": 0.14150448888540268,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6834791004657745,
      "step": 1178
    },
    {
      "completion_length": 108.234375,
      "epoch": 0.7601547388781431,
      "grad_norm": 16.865028381347656,
      "kl": 0.117431640625,
      "learning_rate": 6.199226305609284e-07,
      "loss": 0.0047,
      "reward": 1.5311815738677979,
      "reward_std": 0.058224376291036606,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.531181588768959,
      "step": 1179
    },
    {
      "completion_length": 114.421875,
      "epoch": 0.7607994842037396,
      "grad_norm": 11.300497055053711,
      "kl": 0.1240234375,
      "learning_rate": 6.196002578981302e-07,
      "loss": 0.005,
      "reward": 1.549267292022705,
      "reward_std": 0.11853723600506783,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5492672622203827,
      "step": 1180
    },
    {
      "completion_length": 111.671875,
      "epoch": 0.7614442295293359,
      "grad_norm": 14.773999214172363,
      "kl": 0.11474609375,
      "learning_rate": 6.19277885235332e-07,
      "loss": 0.0046,
      "reward": 1.4692705869674683,
      "reward_std": 0.11159190721809864,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46927061676979065,
      "step": 1181
    },
    {
      "completion_length": 106.671875,
      "epoch": 0.7620889748549323,
      "grad_norm": 12.688148498535156,
      "kl": 0.13671875,
      "learning_rate": 6.189555125725338e-07,
      "loss": 0.0055,
      "reward": 1.6706724166870117,
      "reward_std": 0.1215021051466465,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6706724166870117,
      "step": 1182
    },
    {
      "completion_length": 110.125,
      "epoch": 0.7627337201805287,
      "grad_norm": 24.101341247558594,
      "kl": 0.155029296875,
      "learning_rate": 6.186331399097357e-07,
      "loss": 0.0062,
      "reward": 1.5696587562561035,
      "reward_std": 0.2225308120250702,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5852837711572647,
      "step": 1183
    },
    {
      "completion_length": 107.28125,
      "epoch": 0.7633784655061251,
      "grad_norm": 16.765844345092773,
      "kl": 0.122314453125,
      "learning_rate": 6.183107672469374e-07,
      "loss": 0.0049,
      "reward": 1.7224302291870117,
      "reward_std": 0.08688852190971375,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7224301993846893,
      "step": 1184
    },
    {
      "completion_length": 110.859375,
      "epoch": 0.7640232108317214,
      "grad_norm": 15.215022087097168,
      "kl": 0.154296875,
      "learning_rate": 6.179883945841392e-07,
      "loss": 0.0062,
      "reward": 1.6652826070785522,
      "reward_std": 0.15134353935718536,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6809075474739075,
      "step": 1185
    },
    {
      "completion_length": 109.515625,
      "epoch": 0.7646679561573179,
      "grad_norm": 14.763962745666504,
      "kl": 0.1337890625,
      "learning_rate": 6.17666021921341e-07,
      "loss": 0.0053,
      "reward": 1.4886223673820496,
      "reward_std": 0.14177030697464943,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4886223375797272,
      "step": 1186
    },
    {
      "completion_length": 120.53125,
      "epoch": 0.7653127014829143,
      "grad_norm": 20.856586456298828,
      "kl": 0.11962890625,
      "learning_rate": 6.173436492585429e-07,
      "loss": 0.0048,
      "reward": 1.5948373079299927,
      "reward_std": 0.12407495453953743,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5948372483253479,
      "step": 1187
    },
    {
      "completion_length": 106.53125,
      "epoch": 0.7659574468085106,
      "grad_norm": 7.932638168334961,
      "kl": 0.14404296875,
      "learning_rate": 6.170212765957446e-07,
      "loss": 0.0058,
      "reward": 1.6733270287513733,
      "reward_std": 0.14593126624822617,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6733270585536957,
      "step": 1188
    },
    {
      "completion_length": 122.390625,
      "epoch": 0.7666021921341071,
      "grad_norm": 6.648770332336426,
      "kl": 0.14501953125,
      "learning_rate": 6.166989039329464e-07,
      "loss": 0.0058,
      "reward": 1.4227712154388428,
      "reward_std": 0.11841831728816032,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.42277124524116516,
      "step": 1189
    },
    {
      "completion_length": 109.3125,
      "epoch": 0.7672469374597034,
      "grad_norm": 9.889556884765625,
      "kl": 0.11328125,
      "learning_rate": 6.163765312701482e-07,
      "loss": 0.0045,
      "reward": 1.6376824975013733,
      "reward_std": 0.06098635122179985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6376824975013733,
      "step": 1190
    },
    {
      "completion_length": 114.984375,
      "epoch": 0.7678916827852998,
      "grad_norm": 14.385912895202637,
      "kl": 0.1591796875,
      "learning_rate": 6.160541586073501e-07,
      "loss": 0.0063,
      "reward": 1.550232172012329,
      "reward_std": 0.0959642305970192,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5502321124076843,
      "step": 1191
    },
    {
      "completion_length": 104.5,
      "epoch": 0.7685364281108962,
      "grad_norm": 23.22954559326172,
      "kl": 0.124267578125,
      "learning_rate": 6.157317859445518e-07,
      "loss": 0.005,
      "reward": 1.396329402923584,
      "reward_std": 0.11338374391198158,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.411954402923584,
      "step": 1192
    },
    {
      "completion_length": 111.671875,
      "epoch": 0.7691811734364926,
      "grad_norm": 16.116065979003906,
      "kl": 0.12353515625,
      "learning_rate": 6.154094132817537e-07,
      "loss": 0.0049,
      "reward": 1.7101741433143616,
      "reward_std": 0.083702202886343,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7101741433143616,
      "step": 1193
    },
    {
      "completion_length": 126.359375,
      "epoch": 0.769825918762089,
      "grad_norm": 12.175841331481934,
      "kl": 0.114501953125,
      "learning_rate": 6.150870406189554e-07,
      "loss": 0.0046,
      "reward": 1.7419912219047546,
      "reward_std": 0.08732573315501213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7419912219047546,
      "step": 1194
    },
    {
      "completion_length": 110.546875,
      "epoch": 0.7704706640876854,
      "grad_norm": 13.75151252746582,
      "kl": 0.114013671875,
      "learning_rate": 6.147646679561573e-07,
      "loss": 0.0046,
      "reward": 1.6501139402389526,
      "reward_std": 0.11225607246160507,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6501139998435974,
      "step": 1195
    },
    {
      "completion_length": 106.46875,
      "epoch": 0.7711154094132817,
      "grad_norm": 11.834531784057617,
      "kl": 0.13818359375,
      "learning_rate": 6.144422952933591e-07,
      "loss": 0.0055,
      "reward": 1.676537036895752,
      "reward_std": 0.09139727428555489,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6765370666980743,
      "step": 1196
    },
    {
      "completion_length": 109.140625,
      "epoch": 0.7717601547388782,
      "grad_norm": 29.20878791809082,
      "kl": 0.12548828125,
      "learning_rate": 6.141199226305609e-07,
      "loss": 0.005,
      "reward": 1.5550090074539185,
      "reward_std": 0.13649238646030426,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5706340670585632,
      "step": 1197
    },
    {
      "completion_length": 112.53125,
      "epoch": 0.7724049000644745,
      "grad_norm": 13.043038368225098,
      "kl": 0.140869140625,
      "learning_rate": 6.137975499677626e-07,
      "loss": 0.0056,
      "reward": 1.5729409456253052,
      "reward_std": 0.16874703764915466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5729409903287888,
      "step": 1198
    },
    {
      "completion_length": 106.4375,
      "epoch": 0.7730496453900709,
      "grad_norm": 14.998727798461914,
      "kl": 0.14599609375,
      "learning_rate": 6.134751773049646e-07,
      "loss": 0.0058,
      "reward": 1.5888980627059937,
      "reward_std": 0.12762803584337234,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5888980329036713,
      "step": 1199
    },
    {
      "completion_length": 111.9375,
      "epoch": 0.7736943907156673,
      "grad_norm": 107.7376937866211,
      "kl": 0.13671875,
      "learning_rate": 6.131528046421663e-07,
      "loss": 0.0055,
      "reward": 1.6070385575294495,
      "reward_std": 0.10263358429074287,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6070385873317719,
      "step": 1200
    },
    {
      "completion_length": 113.765625,
      "epoch": 0.7743391360412637,
      "grad_norm": 7.839345455169678,
      "kl": 0.13916015625,
      "learning_rate": 6.128304319793681e-07,
      "loss": 0.0056,
      "reward": 1.6572014093399048,
      "reward_std": 0.11518439650535583,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6572013795375824,
      "step": 1201
    },
    {
      "completion_length": 98.234375,
      "epoch": 0.7749838813668601,
      "grad_norm": 14.737028121948242,
      "kl": 0.1513671875,
      "learning_rate": 6.125080593165699e-07,
      "loss": 0.006,
      "reward": 1.52261883020401,
      "reward_std": 0.15272536873817444,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5382438600063324,
      "step": 1202
    },
    {
      "completion_length": 113.875,
      "epoch": 0.7756286266924565,
      "grad_norm": 8.930920600891113,
      "kl": 0.134033203125,
      "learning_rate": 6.121856866537717e-07,
      "loss": 0.0054,
      "reward": 1.5145543813705444,
      "reward_std": 0.14755171723663807,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.530179426074028,
      "step": 1203
    },
    {
      "completion_length": 109.40625,
      "epoch": 0.7762733720180529,
      "grad_norm": 8.757515907287598,
      "kl": 0.130859375,
      "learning_rate": 6.118633139909736e-07,
      "loss": 0.0052,
      "reward": 1.53077232837677,
      "reward_std": 0.10953427106142044,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5307723581790924,
      "step": 1204
    },
    {
      "completion_length": 113.125,
      "epoch": 0.7769181173436492,
      "grad_norm": 203.08412170410156,
      "kl": 2.7158203125,
      "learning_rate": 6.115409413281754e-07,
      "loss": 0.1085,
      "reward": 1.5865591764450073,
      "reward_std": 0.08946330845355988,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5865591168403625,
      "step": 1205
    },
    {
      "completion_length": 108.046875,
      "epoch": 0.7775628626692457,
      "grad_norm": 11.308333396911621,
      "kl": 0.14013671875,
      "learning_rate": 6.112185686653771e-07,
      "loss": 0.0056,
      "reward": 1.4254971742630005,
      "reward_std": 0.2110433205962181,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4411221593618393,
      "step": 1206
    },
    {
      "completion_length": 106.875,
      "epoch": 0.778207607994842,
      "grad_norm": 13.741293907165527,
      "kl": 0.12646484375,
      "learning_rate": 6.108961960025789e-07,
      "loss": 0.0051,
      "reward": 1.6106538772583008,
      "reward_std": 0.13345135748386383,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6106538474559784,
      "step": 1207
    },
    {
      "completion_length": 101.46875,
      "epoch": 0.7788523533204385,
      "grad_norm": 8.687922477722168,
      "kl": 0.14111328125,
      "learning_rate": 6.105738233397809e-07,
      "loss": 0.0057,
      "reward": 1.4746322631835938,
      "reward_std": 0.19175250083208084,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5058822631835938,
      "step": 1208
    },
    {
      "completion_length": 99.484375,
      "epoch": 0.7794970986460348,
      "grad_norm": 12.14449691772461,
      "kl": 0.15185546875,
      "learning_rate": 6.102514506769826e-07,
      "loss": 0.0061,
      "reward": 1.7666597962379456,
      "reward_std": 0.18449196964502335,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7822847962379456,
      "step": 1209
    },
    {
      "completion_length": 106.03125,
      "epoch": 0.7801418439716312,
      "grad_norm": 22.527015686035156,
      "kl": 0.13916015625,
      "learning_rate": 6.099290780141844e-07,
      "loss": 0.0056,
      "reward": 1.5839348435401917,
      "reward_std": 0.112147755920887,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5839348137378693,
      "step": 1210
    },
    {
      "completion_length": 102.15625,
      "epoch": 0.7807865892972276,
      "grad_norm": 46.15434265136719,
      "kl": 0.1787109375,
      "learning_rate": 6.096067053513861e-07,
      "loss": 0.0072,
      "reward": 1.5750333070755005,
      "reward_std": 0.0655747577548027,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5750332772731781,
      "step": 1211
    },
    {
      "completion_length": 112.359375,
      "epoch": 0.781431334622824,
      "grad_norm": 14.038028717041016,
      "kl": 0.1357421875,
      "learning_rate": 6.092843326885881e-07,
      "loss": 0.0054,
      "reward": 1.416702389717102,
      "reward_std": 0.16470851749181747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41670234501361847,
      "step": 1212
    },
    {
      "completion_length": 93.0,
      "epoch": 0.7820760799484203,
      "grad_norm": 25.67938804626465,
      "kl": 0.15087890625,
      "learning_rate": 6.089619600257898e-07,
      "loss": 0.006,
      "reward": 1.6413110494613647,
      "reward_std": 0.07309877313673496,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.64131098985672,
      "step": 1213
    },
    {
      "completion_length": 105.5625,
      "epoch": 0.7827208252740168,
      "grad_norm": 14.893291473388672,
      "kl": 0.15966796875,
      "learning_rate": 6.086395873629916e-07,
      "loss": 0.0064,
      "reward": 1.4017677307128906,
      "reward_std": 0.2676720768213272,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.4330177456140518,
      "step": 1214
    },
    {
      "completion_length": 93.484375,
      "epoch": 0.7833655705996132,
      "grad_norm": 9.510297775268555,
      "kl": 0.126708984375,
      "learning_rate": 6.083172147001934e-07,
      "loss": 0.0051,
      "reward": 1.4641136527061462,
      "reward_std": 0.11368128657341003,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.464113712310791,
      "step": 1215
    },
    {
      "completion_length": 101.515625,
      "epoch": 0.7840103159252095,
      "grad_norm": 23.989641189575195,
      "kl": 0.1552734375,
      "learning_rate": 6.079948420373952e-07,
      "loss": 0.0062,
      "reward": 1.7286843061447144,
      "reward_std": 0.10229164734482765,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7286842465400696,
      "step": 1216
    },
    {
      "completion_length": 100.609375,
      "epoch": 0.784655061250806,
      "grad_norm": 11.944487571716309,
      "kl": 0.14306640625,
      "learning_rate": 6.07672469374597e-07,
      "loss": 0.0057,
      "reward": 1.5962103009223938,
      "reward_std": 0.11994421109557152,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5962103307247162,
      "step": 1217
    },
    {
      "completion_length": 100.453125,
      "epoch": 0.7852998065764023,
      "grad_norm": 13.68811321258545,
      "kl": 0.141845703125,
      "learning_rate": 6.073500967117989e-07,
      "loss": 0.0057,
      "reward": 1.5690941214561462,
      "reward_std": 0.10337880253791809,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5690941512584686,
      "step": 1218
    },
    {
      "completion_length": 96.4375,
      "epoch": 0.7859445519019987,
      "grad_norm": 22.196964263916016,
      "kl": 0.14306640625,
      "learning_rate": 6.070277240490006e-07,
      "loss": 0.0057,
      "reward": 1.3420570492744446,
      "reward_std": 0.11058967746794224,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.342057004570961,
      "step": 1219
    },
    {
      "completion_length": 97.875,
      "epoch": 0.7865892972275951,
      "grad_norm": 35.983314514160156,
      "kl": 0.1220703125,
      "learning_rate": 6.067053513862024e-07,
      "loss": 0.0049,
      "reward": 1.577008068561554,
      "reward_std": 0.10896754078567028,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5770081877708435,
      "step": 1220
    },
    {
      "completion_length": 104.453125,
      "epoch": 0.7872340425531915,
      "grad_norm": 20.094215393066406,
      "kl": 0.12890625,
      "learning_rate": 6.063829787234043e-07,
      "loss": 0.0051,
      "reward": 1.6798162460327148,
      "reward_std": 0.18578680604696274,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6954412758350372,
      "step": 1221
    },
    {
      "completion_length": 105.859375,
      "epoch": 0.7878787878787878,
      "grad_norm": 21.012859344482422,
      "kl": 0.15576171875,
      "learning_rate": 6.060606060606061e-07,
      "loss": 0.0062,
      "reward": 1.4645494222640991,
      "reward_std": 0.08789615705609322,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4645494073629379,
      "step": 1222
    },
    {
      "completion_length": 97.828125,
      "epoch": 0.7885235332043843,
      "grad_norm": 21.104331970214844,
      "kl": 0.1318359375,
      "learning_rate": 6.057382333978078e-07,
      "loss": 0.0053,
      "reward": 1.5897061228752136,
      "reward_std": 0.1244133822619915,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.589706152677536,
      "step": 1223
    },
    {
      "completion_length": 95.78125,
      "epoch": 0.7891682785299806,
      "grad_norm": 21.754030227661133,
      "kl": 0.122314453125,
      "learning_rate": 6.054158607350097e-07,
      "loss": 0.0049,
      "reward": 1.6213258504867554,
      "reward_std": 0.12604708224534988,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6369507908821106,
      "step": 1224
    },
    {
      "completion_length": 116.484375,
      "epoch": 0.7898130238555771,
      "grad_norm": 10.232507705688477,
      "kl": 0.12841796875,
      "learning_rate": 6.050934880722115e-07,
      "loss": 0.0051,
      "reward": 1.5714877843856812,
      "reward_std": 0.10408592224121094,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5714877992868423,
      "step": 1225
    },
    {
      "completion_length": 105.84375,
      "epoch": 0.7904577691811734,
      "grad_norm": 12.34595775604248,
      "kl": 0.121337890625,
      "learning_rate": 6.047711154094133e-07,
      "loss": 0.0049,
      "reward": 1.571517288684845,
      "reward_std": 0.13167385384440422,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.571517288684845,
      "step": 1226
    },
    {
      "completion_length": 116.328125,
      "epoch": 0.7911025145067698,
      "grad_norm": 22.388225555419922,
      "kl": 0.21240234375,
      "learning_rate": 6.04448742746615e-07,
      "loss": 0.0085,
      "reward": 1.595013439655304,
      "reward_std": 0.1744321957230568,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.610638439655304,
      "step": 1227
    },
    {
      "completion_length": 106.953125,
      "epoch": 0.7917472598323663,
      "grad_norm": 10.770123481750488,
      "kl": 0.138671875,
      "learning_rate": 6.041263700838169e-07,
      "loss": 0.0056,
      "reward": 1.6746158003807068,
      "reward_std": 0.16554439067840576,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6902407705783844,
      "step": 1228
    },
    {
      "completion_length": 98.1875,
      "epoch": 0.7923920051579626,
      "grad_norm": 7.27217960357666,
      "kl": 0.12890625,
      "learning_rate": 6.038039974210186e-07,
      "loss": 0.0052,
      "reward": 1.4847107529640198,
      "reward_std": 0.08111886493861675,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4847107529640198,
      "step": 1229
    },
    {
      "completion_length": 105.53125,
      "epoch": 0.793036750483559,
      "grad_norm": 33.54627990722656,
      "kl": 0.12841796875,
      "learning_rate": 6.034816247582206e-07,
      "loss": 0.0051,
      "reward": 1.6125401258468628,
      "reward_std": 0.18335703760385513,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6125401556491852,
      "step": 1230
    },
    {
      "completion_length": 99.484375,
      "epoch": 0.7936814958091554,
      "grad_norm": 10.74837589263916,
      "kl": 0.2080078125,
      "learning_rate": 6.031592520954223e-07,
      "loss": 0.0083,
      "reward": 1.5733410716056824,
      "reward_std": 0.2276994064450264,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5889661312103271,
      "step": 1231
    },
    {
      "completion_length": 103.171875,
      "epoch": 0.7943262411347518,
      "grad_norm": 14.955659866333008,
      "kl": 0.12353515625,
      "learning_rate": 6.028368794326241e-07,
      "loss": 0.0049,
      "reward": 1.462202548980713,
      "reward_std": 0.20029695332050323,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4778275787830353,
      "step": 1232
    },
    {
      "completion_length": 107.609375,
      "epoch": 0.7949709864603481,
      "grad_norm": 10.545095443725586,
      "kl": 0.13134765625,
      "learning_rate": 6.025145067698258e-07,
      "loss": 0.0052,
      "reward": 1.440338671207428,
      "reward_std": 0.1069004088640213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4403386861085892,
      "step": 1233
    },
    {
      "completion_length": 108.578125,
      "epoch": 0.7956157317859446,
      "grad_norm": 16.158153533935547,
      "kl": 0.135009765625,
      "learning_rate": 6.021921341070278e-07,
      "loss": 0.0054,
      "reward": 1.503720223903656,
      "reward_std": 0.18556904792785645,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.503720223903656,
      "step": 1234
    },
    {
      "completion_length": 95.71875,
      "epoch": 0.7962604771115409,
      "grad_norm": 10.688615798950195,
      "kl": 0.1416015625,
      "learning_rate": 6.018697614442295e-07,
      "loss": 0.0057,
      "reward": 1.4603247046470642,
      "reward_std": 0.11439431831240654,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.460324689745903,
      "step": 1235
    },
    {
      "completion_length": 99.140625,
      "epoch": 0.7969052224371374,
      "grad_norm": 100.63502502441406,
      "kl": 0.129638671875,
      "learning_rate": 6.015473887814313e-07,
      "loss": 0.0052,
      "reward": 1.771046221256256,
      "reward_std": 0.07787198200821877,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7710461616516113,
      "step": 1236
    },
    {
      "completion_length": 106.015625,
      "epoch": 0.7975499677627337,
      "grad_norm": 8.697854995727539,
      "kl": 0.1162109375,
      "learning_rate": 6.012250161186331e-07,
      "loss": 0.0046,
      "reward": 1.5569599866867065,
      "reward_std": 0.1632199063897133,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5569599568843842,
      "step": 1237
    },
    {
      "completion_length": 104.859375,
      "epoch": 0.7981947130883301,
      "grad_norm": 14.559203147888184,
      "kl": 0.14697265625,
      "learning_rate": 6.00902643455835e-07,
      "loss": 0.0059,
      "reward": 1.6990448832511902,
      "reward_std": 0.177838034927845,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6990448534488678,
      "step": 1238
    },
    {
      "completion_length": 95.234375,
      "epoch": 0.7988394584139265,
      "grad_norm": 47.494850158691406,
      "kl": 0.16357421875,
      "learning_rate": 6.005802707930367e-07,
      "loss": 0.0065,
      "reward": 1.5274017453193665,
      "reward_std": 0.11369790881872177,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5274017453193665,
      "step": 1239
    },
    {
      "completion_length": 100.96875,
      "epoch": 0.7994842037395229,
      "grad_norm": 20.095169067382812,
      "kl": 0.113037109375,
      "learning_rate": 6.002578981302386e-07,
      "loss": 0.0045,
      "reward": 1.5768287777900696,
      "reward_std": 0.15517547726631165,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5768287479877472,
      "step": 1240
    },
    {
      "completion_length": 97.265625,
      "epoch": 0.8001289490651193,
      "grad_norm": 16.03243064880371,
      "kl": 0.1318359375,
      "learning_rate": 5.999355254674403e-07,
      "loss": 0.0053,
      "reward": 1.51595139503479,
      "reward_std": 0.14299090951681137,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5159513652324677,
      "step": 1241
    },
    {
      "completion_length": 103.953125,
      "epoch": 0.8007736943907157,
      "grad_norm": 10.533329010009766,
      "kl": 0.115478515625,
      "learning_rate": 5.996131528046421e-07,
      "loss": 0.0046,
      "reward": 1.4868841767311096,
      "reward_std": 0.18622149527072906,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48688413202762604,
      "step": 1242
    },
    {
      "completion_length": 94.5625,
      "epoch": 0.8014184397163121,
      "grad_norm": 8.86988353729248,
      "kl": 0.15185546875,
      "learning_rate": 5.99290780141844e-07,
      "loss": 0.0061,
      "reward": 1.5329524874687195,
      "reward_std": 0.216628547757864,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5485774874687195,
      "step": 1243
    },
    {
      "completion_length": 95.078125,
      "epoch": 0.8020631850419084,
      "grad_norm": 9.795210838317871,
      "kl": 0.12841796875,
      "learning_rate": 5.989684074790458e-07,
      "loss": 0.0051,
      "reward": 1.6223825812339783,
      "reward_std": 0.1345507875084877,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6223825961351395,
      "step": 1244
    },
    {
      "completion_length": 100.859375,
      "epoch": 0.8027079303675049,
      "grad_norm": 32.753726959228516,
      "kl": 0.125732421875,
      "learning_rate": 5.986460348162475e-07,
      "loss": 0.005,
      "reward": 1.5496293306350708,
      "reward_std": 0.09532084688544273,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.549629271030426,
      "step": 1245
    },
    {
      "completion_length": 98.875,
      "epoch": 0.8033526756931012,
      "grad_norm": 21.810379028320312,
      "kl": 0.1298828125,
      "learning_rate": 5.983236621534494e-07,
      "loss": 0.0052,
      "reward": 1.3395966291427612,
      "reward_std": 0.09464257955551147,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33959658443927765,
      "step": 1246
    },
    {
      "completion_length": 85.375,
      "epoch": 0.8039974210186976,
      "grad_norm": 21.297460556030273,
      "kl": 0.14599609375,
      "learning_rate": 5.980012894906512e-07,
      "loss": 0.0058,
      "reward": 1.7041836380958557,
      "reward_std": 0.10666557401418686,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7041836380958557,
      "step": 1247
    },
    {
      "completion_length": 91.0625,
      "epoch": 0.804642166344294,
      "grad_norm": 13.005825996398926,
      "kl": 0.14453125,
      "learning_rate": 5.97678916827853e-07,
      "loss": 0.0058,
      "reward": 1.7234504222869873,
      "reward_std": 0.1442631632089615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7234503924846649,
      "step": 1248
    },
    {
      "completion_length": 103.765625,
      "epoch": 0.8052869116698904,
      "grad_norm": 41.9098014831543,
      "kl": 0.1171875,
      "learning_rate": 5.973565441650548e-07,
      "loss": 0.0047,
      "reward": 1.4872705340385437,
      "reward_std": 0.17103341966867447,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4872705489397049,
      "step": 1249
    },
    {
      "completion_length": 95.15625,
      "epoch": 0.8059316569954867,
      "grad_norm": 34.45671081542969,
      "kl": 0.13330078125,
      "learning_rate": 5.970341715022566e-07,
      "loss": 0.0053,
      "reward": 1.6393902897834778,
      "reward_std": 0.14729318767786026,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6393902599811554,
      "step": 1250
    },
    {
      "completion_length": 94.25,
      "epoch": 0.8065764023210832,
      "grad_norm": 14.370905876159668,
      "kl": 0.13720703125,
      "learning_rate": 5.967117988394584e-07,
      "loss": 0.0055,
      "reward": 1.5178990960121155,
      "reward_std": 0.07636865600943565,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5178991109132767,
      "step": 1251
    },
    {
      "completion_length": 88.359375,
      "epoch": 0.8072211476466795,
      "grad_norm": 47.83586120605469,
      "kl": 0.14111328125,
      "learning_rate": 5.963894261766603e-07,
      "loss": 0.0057,
      "reward": 1.5101807117462158,
      "reward_std": 0.1659940369427204,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5101807117462158,
      "step": 1252
    },
    {
      "completion_length": 92.328125,
      "epoch": 0.807865892972276,
      "grad_norm": 17.251609802246094,
      "kl": 0.15087890625,
      "learning_rate": 5.96067053513862e-07,
      "loss": 0.006,
      "reward": 1.641158401966095,
      "reward_std": 0.17022105678915977,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.656783401966095,
      "step": 1253
    },
    {
      "completion_length": 85.4375,
      "epoch": 0.8085106382978723,
      "grad_norm": 9.6865873336792,
      "kl": 0.173828125,
      "learning_rate": 5.957446808510638e-07,
      "loss": 0.0069,
      "reward": 1.5700511932373047,
      "reward_std": 0.11514360085129738,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5700511932373047,
      "step": 1254
    },
    {
      "completion_length": 94.03125,
      "epoch": 0.8091553836234687,
      "grad_norm": 21.811676025390625,
      "kl": 0.122802734375,
      "learning_rate": 5.954223081882655e-07,
      "loss": 0.0049,
      "reward": 1.520093321800232,
      "reward_std": 0.16418229416012764,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5200933665037155,
      "step": 1255
    },
    {
      "completion_length": 85.9375,
      "epoch": 0.8098001289490652,
      "grad_norm": 59.537200927734375,
      "kl": 0.1484375,
      "learning_rate": 5.950999355254675e-07,
      "loss": 0.0059,
      "reward": 1.5090277194976807,
      "reward_std": 0.1114044338464737,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5090276747941971,
      "step": 1256
    },
    {
      "completion_length": 92.640625,
      "epoch": 0.8104448742746615,
      "grad_norm": 50.82180404663086,
      "kl": 0.15283203125,
      "learning_rate": 5.947775628626692e-07,
      "loss": 0.0061,
      "reward": 1.3142651319503784,
      "reward_std": 0.037534992210567,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.31426508724689484,
      "step": 1257
    },
    {
      "completion_length": 83.796875,
      "epoch": 0.8110896196002579,
      "grad_norm": 20.525754928588867,
      "kl": 0.1259765625,
      "learning_rate": 5.94455190199871e-07,
      "loss": 0.005,
      "reward": 1.6182879209518433,
      "reward_std": 0.1074063628911972,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6182878911495209,
      "step": 1258
    },
    {
      "completion_length": 87.0,
      "epoch": 0.8117343649258543,
      "grad_norm": 9.627203941345215,
      "kl": 0.11865234375,
      "learning_rate": 5.941328175370728e-07,
      "loss": 0.0047,
      "reward": 1.5465694665908813,
      "reward_std": 0.09133465215563774,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.546569436788559,
      "step": 1259
    },
    {
      "completion_length": 99.09375,
      "epoch": 0.8123791102514507,
      "grad_norm": 22.241708755493164,
      "kl": 0.13037109375,
      "learning_rate": 5.938104448742747e-07,
      "loss": 0.0052,
      "reward": 1.5899001359939575,
      "reward_std": 0.15345070511102676,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6055251061916351,
      "step": 1260
    },
    {
      "completion_length": 96.40625,
      "epoch": 0.813023855577047,
      "grad_norm": 9.629485130310059,
      "kl": 0.1455078125,
      "learning_rate": 5.934880722114764e-07,
      "loss": 0.0058,
      "reward": 1.6390262842178345,
      "reward_std": 0.16355416178703308,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6390263140201569,
      "step": 1261
    },
    {
      "completion_length": 91.40625,
      "epoch": 0.8136686009026435,
      "grad_norm": 39.97684097290039,
      "kl": 0.189453125,
      "learning_rate": 5.931656995486783e-07,
      "loss": 0.0076,
      "reward": 1.5222207903862,
      "reward_std": 0.10405867174267769,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5222208201885223,
      "step": 1262
    },
    {
      "completion_length": 86.65625,
      "epoch": 0.8143133462282398,
      "grad_norm": 13.295222282409668,
      "kl": 0.114990234375,
      "learning_rate": 5.9284332688588e-07,
      "loss": 0.0046,
      "reward": 1.6988143920898438,
      "reward_std": 0.12119985744357109,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6988143026828766,
      "step": 1263
    },
    {
      "completion_length": 99.734375,
      "epoch": 0.8149580915538363,
      "grad_norm": 25.42751693725586,
      "kl": 0.17138671875,
      "learning_rate": 5.925209542230819e-07,
      "loss": 0.0069,
      "reward": 1.5959572792053223,
      "reward_std": 0.10522185638546944,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5959572196006775,
      "step": 1264
    },
    {
      "completion_length": 101.1875,
      "epoch": 0.8156028368794326,
      "grad_norm": 12.357390403747559,
      "kl": 0.19189453125,
      "learning_rate": 5.921985815602837e-07,
      "loss": 0.0077,
      "reward": 1.6559144854545593,
      "reward_std": 0.1504291370511055,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6559144854545593,
      "step": 1265
    },
    {
      "completion_length": 98.828125,
      "epoch": 0.816247582205029,
      "grad_norm": 22.642732620239258,
      "kl": 0.12890625,
      "learning_rate": 5.918762088974855e-07,
      "loss": 0.0052,
      "reward": 1.5548232197761536,
      "reward_std": 0.10006397217512131,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5548232197761536,
      "step": 1266
    },
    {
      "completion_length": 97.125,
      "epoch": 0.8168923275306254,
      "grad_norm": 11.454401969909668,
      "kl": 0.14501953125,
      "learning_rate": 5.915538362346872e-07,
      "loss": 0.0058,
      "reward": 1.6567357182502747,
      "reward_std": 0.09761639684438705,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6567357182502747,
      "step": 1267
    },
    {
      "completion_length": 100.59375,
      "epoch": 0.8175370728562218,
      "grad_norm": 11.95074462890625,
      "kl": 0.1357421875,
      "learning_rate": 5.912314635718891e-07,
      "loss": 0.0054,
      "reward": 1.6781417727470398,
      "reward_std": 0.0793631412088871,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6781417727470398,
      "step": 1268
    },
    {
      "completion_length": 94.03125,
      "epoch": 0.8181818181818182,
      "grad_norm": 23.272687911987305,
      "kl": 0.1201171875,
      "learning_rate": 5.909090909090909e-07,
      "loss": 0.0048,
      "reward": 1.6391507387161255,
      "reward_std": 0.1269705779850483,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6547757685184479,
      "step": 1269
    },
    {
      "completion_length": 95.546875,
      "epoch": 0.8188265635074146,
      "grad_norm": 14.534205436706543,
      "kl": 0.109130859375,
      "learning_rate": 5.905867182462927e-07,
      "loss": 0.0044,
      "reward": 1.5957148671150208,
      "reward_std": 0.09973860159516335,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5957148522138596,
      "step": 1270
    },
    {
      "completion_length": 94.78125,
      "epoch": 0.819471308833011,
      "grad_norm": 14.054845809936523,
      "kl": 0.128173828125,
      "learning_rate": 5.902643455834945e-07,
      "loss": 0.0051,
      "reward": 1.5978807210922241,
      "reward_std": 0.09598659723997116,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5978806614875793,
      "step": 1271
    },
    {
      "completion_length": 94.296875,
      "epoch": 0.8201160541586073,
      "grad_norm": 20.644716262817383,
      "kl": 0.1259765625,
      "learning_rate": 5.899419729206963e-07,
      "loss": 0.0051,
      "reward": 1.4058359265327454,
      "reward_std": 0.11267614737153053,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4058358669281006,
      "step": 1272
    },
    {
      "completion_length": 91.78125,
      "epoch": 0.8207607994842038,
      "grad_norm": 25.96257209777832,
      "kl": 0.11767578125,
      "learning_rate": 5.896196002578981e-07,
      "loss": 0.0047,
      "reward": 1.6652281284332275,
      "reward_std": 0.08674262091517448,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6652281284332275,
      "step": 1273
    },
    {
      "completion_length": 99.078125,
      "epoch": 0.8214055448098001,
      "grad_norm": 12.55455207824707,
      "kl": 0.1318359375,
      "learning_rate": 5.892972275951e-07,
      "loss": 0.0053,
      "reward": 1.500665009021759,
      "reward_std": 0.14083610475063324,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.500665009021759,
      "step": 1274
    },
    {
      "completion_length": 103.890625,
      "epoch": 0.8220502901353965,
      "grad_norm": 21.38951301574707,
      "kl": 0.1318359375,
      "learning_rate": 5.889748549323017e-07,
      "loss": 0.0053,
      "reward": 1.6267691254615784,
      "reward_std": 0.17314469069242477,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6423941254615784,
      "step": 1275
    },
    {
      "completion_length": 88.046875,
      "epoch": 0.8226950354609929,
      "grad_norm": 20.793231964111328,
      "kl": 0.1357421875,
      "learning_rate": 5.886524822695035e-07,
      "loss": 0.0054,
      "reward": 1.661752164363861,
      "reward_std": 0.11195382103323936,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6617521345615387,
      "step": 1276
    },
    {
      "completion_length": 98.78125,
      "epoch": 0.8233397807865893,
      "grad_norm": 13.758505821228027,
      "kl": 0.12353515625,
      "learning_rate": 5.883301096067053e-07,
      "loss": 0.0049,
      "reward": 1.5527395009994507,
      "reward_std": 0.05753506347537041,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5527395904064178,
      "step": 1277
    },
    {
      "completion_length": 106.171875,
      "epoch": 0.8239845261121856,
      "grad_norm": 11.539607048034668,
      "kl": 0.16064453125,
      "learning_rate": 5.880077369439072e-07,
      "loss": 0.0064,
      "reward": 1.5900872349739075,
      "reward_std": 0.10428940504789352,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5900872051715851,
      "step": 1278
    },
    {
      "completion_length": 94.890625,
      "epoch": 0.8246292714377821,
      "grad_norm": 11.546834945678711,
      "kl": 0.111083984375,
      "learning_rate": 5.876853642811089e-07,
      "loss": 0.0044,
      "reward": 1.566346824169159,
      "reward_std": 0.05556929484009743,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5663468241691589,
      "step": 1279
    },
    {
      "completion_length": 104.25,
      "epoch": 0.8252740167633784,
      "grad_norm": 13.243741035461426,
      "kl": 0.13623046875,
      "learning_rate": 5.873629916183107e-07,
      "loss": 0.0054,
      "reward": 1.7236492037773132,
      "reward_std": 0.07918451726436615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7236492037773132,
      "step": 1280
    },
    {
      "completion_length": 100.078125,
      "epoch": 0.8259187620889749,
      "grad_norm": 23.4504451751709,
      "kl": 0.1435546875,
      "learning_rate": 5.870406189555126e-07,
      "loss": 0.0057,
      "reward": 1.6817771196365356,
      "reward_std": 0.10275669768452644,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6817770898342133,
      "step": 1281
    },
    {
      "completion_length": 101.328125,
      "epoch": 0.8265635074145713,
      "grad_norm": 10.985300064086914,
      "kl": 0.13134765625,
      "learning_rate": 5.867182462927144e-07,
      "loss": 0.0053,
      "reward": 1.5396443605422974,
      "reward_std": 0.08468784764409065,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5396443605422974,
      "step": 1282
    },
    {
      "completion_length": 100.78125,
      "epoch": 0.8272082527401676,
      "grad_norm": 14.259724617004395,
      "kl": 0.145263671875,
      "learning_rate": 5.863958736299161e-07,
      "loss": 0.0058,
      "reward": 1.3875265717506409,
      "reward_std": 0.10882111266255379,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4031515270471573,
      "step": 1283
    },
    {
      "completion_length": 96.875,
      "epoch": 0.8278529980657641,
      "grad_norm": 13.990731239318848,
      "kl": 0.13916015625,
      "learning_rate": 5.86073500967118e-07,
      "loss": 0.0056,
      "reward": 1.6068326830863953,
      "reward_std": 0.1714675948023796,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6224577128887177,
      "step": 1284
    },
    {
      "completion_length": 108.359375,
      "epoch": 0.8284977433913604,
      "grad_norm": 11.542860984802246,
      "kl": 0.150390625,
      "learning_rate": 5.857511283043197e-07,
      "loss": 0.006,
      "reward": 1.5005378723144531,
      "reward_std": 0.07108896225690842,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5005379021167755,
      "step": 1285
    },
    {
      "completion_length": 100.796875,
      "epoch": 0.8291424887169568,
      "grad_norm": 12.178953170776367,
      "kl": 0.16552734375,
      "learning_rate": 5.854287556415216e-07,
      "loss": 0.0066,
      "reward": 1.4452755451202393,
      "reward_std": 0.2179737687110901,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.46090058982372284,
      "step": 1286
    },
    {
      "completion_length": 97.609375,
      "epoch": 0.8297872340425532,
      "grad_norm": 43.12259292602539,
      "kl": 0.123291015625,
      "learning_rate": 5.851063829787234e-07,
      "loss": 0.0049,
      "reward": 1.4171838164329529,
      "reward_std": 0.15592122077941895,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4171838164329529,
      "step": 1287
    },
    {
      "completion_length": 102.0,
      "epoch": 0.8304319793681496,
      "grad_norm": 17.38705062866211,
      "kl": 0.1328125,
      "learning_rate": 5.847840103159252e-07,
      "loss": 0.0053,
      "reward": 1.6413039565086365,
      "reward_std": 0.09573720023036003,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6413039267063141,
      "step": 1288
    },
    {
      "completion_length": 99.71875,
      "epoch": 0.8310767246937459,
      "grad_norm": 12.098993301391602,
      "kl": 0.140625,
      "learning_rate": 5.844616376531269e-07,
      "loss": 0.0056,
      "reward": 1.4964683651924133,
      "reward_std": 0.0922405980527401,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49646835029125214,
      "step": 1289
    },
    {
      "completion_length": 95.625,
      "epoch": 0.8317214700193424,
      "grad_norm": 24.151933670043945,
      "kl": 0.125,
      "learning_rate": 5.841392649903289e-07,
      "loss": 0.005,
      "reward": 1.5137308239936829,
      "reward_std": 0.13668807595968246,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5137307941913605,
      "step": 1290
    },
    {
      "completion_length": 103.234375,
      "epoch": 0.8323662153449387,
      "grad_norm": 17.465185165405273,
      "kl": 0.116943359375,
      "learning_rate": 5.838168923275306e-07,
      "loss": 0.0047,
      "reward": 1.5166469812393188,
      "reward_std": 0.08108877390623093,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5166469812393188,
      "step": 1291
    },
    {
      "completion_length": 107.65625,
      "epoch": 0.8330109606705351,
      "grad_norm": 19.077016830444336,
      "kl": 0.14892578125,
      "learning_rate": 5.834945196647324e-07,
      "loss": 0.006,
      "reward": 1.5799004435539246,
      "reward_std": 0.10533102601766586,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5799004435539246,
      "step": 1292
    },
    {
      "completion_length": 113.484375,
      "epoch": 0.8336557059961315,
      "grad_norm": 8.121297836303711,
      "kl": 0.1162109375,
      "learning_rate": 5.831721470019342e-07,
      "loss": 0.0047,
      "reward": 1.626124620437622,
      "reward_std": 0.05737309157848358,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6261245012283325,
      "step": 1293
    },
    {
      "completion_length": 113.484375,
      "epoch": 0.8343004513217279,
      "grad_norm": 15.856319427490234,
      "kl": 0.119140625,
      "learning_rate": 5.828497743391361e-07,
      "loss": 0.0048,
      "reward": 1.5811055302619934,
      "reward_std": 0.09756194055080414,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5811055451631546,
      "step": 1294
    },
    {
      "completion_length": 118.125,
      "epoch": 0.8349451966473244,
      "grad_norm": 28.771743774414062,
      "kl": 0.112060546875,
      "learning_rate": 5.825274016763378e-07,
      "loss": 0.0045,
      "reward": 1.590721607208252,
      "reward_std": 0.08623911440372467,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5907216668128967,
      "step": 1295
    },
    {
      "completion_length": 99.09375,
      "epoch": 0.8355899419729207,
      "grad_norm": 9.904467582702637,
      "kl": 0.146484375,
      "learning_rate": 5.822050290135397e-07,
      "loss": 0.0059,
      "reward": 1.6357603073120117,
      "reward_std": 0.10336859151721,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6357603967189789,
      "step": 1296
    },
    {
      "completion_length": 107.90625,
      "epoch": 0.8362346872985171,
      "grad_norm": 9.767264366149902,
      "kl": 0.11328125,
      "learning_rate": 5.818826563507414e-07,
      "loss": 0.0045,
      "reward": 1.4161508083343506,
      "reward_std": 0.09276717901229858,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4161507785320282,
      "step": 1297
    },
    {
      "completion_length": 106.53125,
      "epoch": 0.8368794326241135,
      "grad_norm": 10.562418937683105,
      "kl": 0.143798828125,
      "learning_rate": 5.815602836879432e-07,
      "loss": 0.0058,
      "reward": 1.7008376717567444,
      "reward_std": 0.1603548526763916,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7008376121520996,
      "step": 1298
    },
    {
      "completion_length": 115.5625,
      "epoch": 0.8375241779497099,
      "grad_norm": 11.865731239318848,
      "kl": 0.138671875,
      "learning_rate": 5.81237911025145e-07,
      "loss": 0.0055,
      "reward": 1.5670779943466187,
      "reward_std": 0.12002968043088913,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.567078024148941,
      "step": 1299
    },
    {
      "completion_length": 106.875,
      "epoch": 0.8381689232753062,
      "grad_norm": 7.02225923538208,
      "kl": 0.1357421875,
      "learning_rate": 5.809155383623469e-07,
      "loss": 0.0054,
      "reward": 1.5539129376411438,
      "reward_std": 0.06455387361347675,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5539128929376602,
      "step": 1300
    },
    {
      "completion_length": 114.25,
      "epoch": 0.8388136686009027,
      "grad_norm": 15.595966339111328,
      "kl": 0.1181640625,
      "learning_rate": 5.805931656995486e-07,
      "loss": 0.0047,
      "reward": 1.5586952567100525,
      "reward_std": 0.10112509876489639,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5586952567100525,
      "step": 1301
    },
    {
      "completion_length": 108.640625,
      "epoch": 0.839458413926499,
      "grad_norm": 8.791936874389648,
      "kl": 0.115234375,
      "learning_rate": 5.802707930367504e-07,
      "loss": 0.0046,
      "reward": 1.6700953841209412,
      "reward_std": 0.06693680211901665,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6700953841209412,
      "step": 1302
    },
    {
      "completion_length": 102.78125,
      "epoch": 0.8401031592520954,
      "grad_norm": 15.279678344726562,
      "kl": 0.11328125,
      "learning_rate": 5.799484203739523e-07,
      "loss": 0.0045,
      "reward": 1.668242335319519,
      "reward_std": 0.061969924718141556,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6682423055171967,
      "step": 1303
    },
    {
      "completion_length": 110.9375,
      "epoch": 0.8407479045776918,
      "grad_norm": 34.09355163574219,
      "kl": 0.15673828125,
      "learning_rate": 5.796260477111541e-07,
      "loss": 0.0063,
      "reward": 1.6093423962593079,
      "reward_std": 0.10130022093653679,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6093423962593079,
      "step": 1304
    },
    {
      "completion_length": 100.296875,
      "epoch": 0.8413926499032882,
      "grad_norm": 16.117502212524414,
      "kl": 0.1552734375,
      "learning_rate": 5.793036750483558e-07,
      "loss": 0.0062,
      "reward": 1.7314233183860779,
      "reward_std": 0.09649763628840446,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7314233183860779,
      "step": 1305
    },
    {
      "completion_length": 100.125,
      "epoch": 0.8420373952288845,
      "grad_norm": 10.007387161254883,
      "kl": 0.13623046875,
      "learning_rate": 5.789813023855577e-07,
      "loss": 0.0054,
      "reward": 1.7936328649520874,
      "reward_std": 0.07578852772712708,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7936329245567322,
      "step": 1306
    },
    {
      "completion_length": 98.40625,
      "epoch": 0.842682140554481,
      "grad_norm": 32.72197723388672,
      "kl": 0.125732421875,
      "learning_rate": 5.786589297227595e-07,
      "loss": 0.005,
      "reward": 1.3983741998672485,
      "reward_std": 0.09278368949890137,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3983742743730545,
      "step": 1307
    },
    {
      "completion_length": 106.078125,
      "epoch": 0.8433268858800773,
      "grad_norm": 10.910943031311035,
      "kl": 0.18505859375,
      "learning_rate": 5.783365570599613e-07,
      "loss": 0.0074,
      "reward": 1.645573914051056,
      "reward_std": 0.14889878779649734,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6611989140510559,
      "step": 1308
    },
    {
      "completion_length": 91.109375,
      "epoch": 0.8439716312056738,
      "grad_norm": 10.358012199401855,
      "kl": 0.1220703125,
      "learning_rate": 5.780141843971631e-07,
      "loss": 0.0049,
      "reward": 1.4291784167289734,
      "reward_std": 0.10380928218364716,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4291783422231674,
      "step": 1309
    },
    {
      "completion_length": 98.453125,
      "epoch": 0.8446163765312702,
      "grad_norm": 21.05621910095215,
      "kl": 0.11865234375,
      "learning_rate": 5.776918117343649e-07,
      "loss": 0.0047,
      "reward": 1.5606682300567627,
      "reward_std": 0.09428859874606133,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5606682300567627,
      "step": 1310
    },
    {
      "completion_length": 112.734375,
      "epoch": 0.8452611218568665,
      "grad_norm": 14.857038497924805,
      "kl": 0.12060546875,
      "learning_rate": 5.773694390715666e-07,
      "loss": 0.0048,
      "reward": 1.736677885055542,
      "reward_std": 0.07061555609107018,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7366778552532196,
      "step": 1311
    },
    {
      "completion_length": 105.25,
      "epoch": 0.845905867182463,
      "grad_norm": 16.779647827148438,
      "kl": 0.1328125,
      "learning_rate": 5.770470664087686e-07,
      "loss": 0.0053,
      "reward": 1.5583295822143555,
      "reward_std": 0.08000925555825233,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5583296418190002,
      "step": 1312
    },
    {
      "completion_length": 101.453125,
      "epoch": 0.8465506125080593,
      "grad_norm": 26.074615478515625,
      "kl": 0.14404296875,
      "learning_rate": 5.767246937459703e-07,
      "loss": 0.0058,
      "reward": 1.5779609084129333,
      "reward_std": 0.09761778265237808,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.577960878610611,
      "step": 1313
    },
    {
      "completion_length": 101.40625,
      "epoch": 0.8471953578336557,
      "grad_norm": 14.144932746887207,
      "kl": 0.1474609375,
      "learning_rate": 5.764023210831721e-07,
      "loss": 0.0059,
      "reward": 1.6106492280960083,
      "reward_std": 0.09668288752436638,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6106491684913635,
      "step": 1314
    },
    {
      "completion_length": 95.40625,
      "epoch": 0.8478401031592521,
      "grad_norm": 24.63298988342285,
      "kl": 0.10791015625,
      "learning_rate": 5.760799484203739e-07,
      "loss": 0.0043,
      "reward": 1.626054286956787,
      "reward_std": 0.1346743032336235,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6260542571544647,
      "step": 1315
    },
    {
      "completion_length": 97.015625,
      "epoch": 0.8484848484848485,
      "grad_norm": 13.633806228637695,
      "kl": 0.13232421875,
      "learning_rate": 5.757575757575758e-07,
      "loss": 0.0053,
      "reward": 1.6472105979919434,
      "reward_std": 0.10808993130922318,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6472106575965881,
      "step": 1316
    },
    {
      "completion_length": 100.046875,
      "epoch": 0.8491295938104448,
      "grad_norm": 31.76630210876465,
      "kl": 0.169921875,
      "learning_rate": 5.754352030947775e-07,
      "loss": 0.0068,
      "reward": 1.5727328062057495,
      "reward_std": 0.1476679891347885,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5883578211069107,
      "step": 1317
    },
    {
      "completion_length": 110.453125,
      "epoch": 0.8497743391360413,
      "grad_norm": 12.650569915771484,
      "kl": 0.1220703125,
      "learning_rate": 5.751128304319794e-07,
      "loss": 0.0049,
      "reward": 1.6151057481765747,
      "reward_std": 0.07021662034094334,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6151057481765747,
      "step": 1318
    },
    {
      "completion_length": 107.203125,
      "epoch": 0.8504190844616376,
      "grad_norm": 26.324243545532227,
      "kl": 0.11181640625,
      "learning_rate": 5.747904577691811e-07,
      "loss": 0.0045,
      "reward": 1.6326099038124084,
      "reward_std": 0.08545324578881264,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6326098740100861,
      "step": 1319
    },
    {
      "completion_length": 98.984375,
      "epoch": 0.851063829787234,
      "grad_norm": 23.667394638061523,
      "kl": 0.10888671875,
      "learning_rate": 5.74468085106383e-07,
      "loss": 0.0044,
      "reward": 1.568135380744934,
      "reward_std": 0.15582210570573807,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5681354403495789,
      "step": 1320
    },
    {
      "completion_length": 102.15625,
      "epoch": 0.8517085751128304,
      "grad_norm": 31.68404197692871,
      "kl": 0.124755859375,
      "learning_rate": 5.741457124435847e-07,
      "loss": 0.005,
      "reward": 1.7288174033164978,
      "reward_std": 0.07873563468456268,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.728817343711853,
      "step": 1321
    },
    {
      "completion_length": 99.9375,
      "epoch": 0.8523533204384268,
      "grad_norm": 78.26886749267578,
      "kl": 0.17138671875,
      "learning_rate": 5.738233397807866e-07,
      "loss": 0.0068,
      "reward": 1.5360566973686218,
      "reward_std": 0.19280319660902023,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5516817271709442,
      "step": 1322
    },
    {
      "completion_length": 98.46875,
      "epoch": 0.8529980657640233,
      "grad_norm": 26.776473999023438,
      "kl": 0.134765625,
      "learning_rate": 5.735009671179883e-07,
      "loss": 0.0054,
      "reward": 1.6314437985420227,
      "reward_std": 0.15900789201259613,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6470687091350555,
      "step": 1323
    },
    {
      "completion_length": 95.078125,
      "epoch": 0.8536428110896196,
      "grad_norm": 29.28158950805664,
      "kl": 0.111572265625,
      "learning_rate": 5.731785944551901e-07,
      "loss": 0.0045,
      "reward": 1.4915387034416199,
      "reward_std": 0.062356844544410706,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49153879284858704,
      "step": 1324
    },
    {
      "completion_length": 108.453125,
      "epoch": 0.854287556415216,
      "grad_norm": 14.453795433044434,
      "kl": 0.12255859375,
      "learning_rate": 5.72856221792392e-07,
      "loss": 0.0049,
      "reward": 1.4318147897720337,
      "reward_std": 0.11395280063152313,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4318147897720337,
      "step": 1325
    },
    {
      "completion_length": 94.96875,
      "epoch": 0.8549323017408124,
      "grad_norm": 23.957849502563477,
      "kl": 0.121826171875,
      "learning_rate": 5.725338491295938e-07,
      "loss": 0.0049,
      "reward": 1.5138567686080933,
      "reward_std": 0.23028697073459625,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5451066493988037,
      "step": 1326
    },
    {
      "completion_length": 103.640625,
      "epoch": 0.8555770470664088,
      "grad_norm": 17.610496520996094,
      "kl": 0.1201171875,
      "learning_rate": 5.722114764667955e-07,
      "loss": 0.0048,
      "reward": 1.4641339778900146,
      "reward_std": 0.10243158787488937,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46413394808769226,
      "step": 1327
    },
    {
      "completion_length": 96.515625,
      "epoch": 0.8562217923920051,
      "grad_norm": 9.11087417602539,
      "kl": 0.1357421875,
      "learning_rate": 5.718891038039974e-07,
      "loss": 0.0054,
      "reward": 1.6933382749557495,
      "reward_std": 0.10557930544018745,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6933382451534271,
      "step": 1328
    },
    {
      "completion_length": 107.765625,
      "epoch": 0.8568665377176016,
      "grad_norm": 27.55734634399414,
      "kl": 0.140625,
      "learning_rate": 5.715667311411992e-07,
      "loss": 0.0056,
      "reward": 1.480468213558197,
      "reward_std": 0.11969191953539848,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.480468213558197,
      "step": 1329
    },
    {
      "completion_length": 101.8125,
      "epoch": 0.8575112830431979,
      "grad_norm": 10.899490356445312,
      "kl": 0.1455078125,
      "learning_rate": 5.71244358478401e-07,
      "loss": 0.0058,
      "reward": 1.4227421879768372,
      "reward_std": 0.1466749757528305,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.43836718797683716,
      "step": 1330
    },
    {
      "completion_length": 107.53125,
      "epoch": 0.8581560283687943,
      "grad_norm": 17.06650733947754,
      "kl": 0.11669921875,
      "learning_rate": 5.709219858156028e-07,
      "loss": 0.0047,
      "reward": 1.5130248665809631,
      "reward_std": 0.13780354335904121,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5286498665809631,
      "step": 1331
    },
    {
      "completion_length": 103.296875,
      "epoch": 0.8588007736943907,
      "grad_norm": 44.2659797668457,
      "kl": 0.119140625,
      "learning_rate": 5.705996131528046e-07,
      "loss": 0.0048,
      "reward": 1.6124387383460999,
      "reward_std": 0.11327242106199265,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6124387979507446,
      "step": 1332
    },
    {
      "completion_length": 93.625,
      "epoch": 0.8594455190199871,
      "grad_norm": 119.70891571044922,
      "kl": 0.13330078125,
      "learning_rate": 5.702772404900064e-07,
      "loss": 0.0053,
      "reward": 1.6062958240509033,
      "reward_std": 0.04072526656091213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6062957644462585,
      "step": 1333
    },
    {
      "completion_length": 105.734375,
      "epoch": 0.8600902643455834,
      "grad_norm": 35.184932708740234,
      "kl": 0.119384765625,
      "learning_rate": 5.699548678272083e-07,
      "loss": 0.0048,
      "reward": 1.557755708694458,
      "reward_std": 0.07671729475259781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5577556788921356,
      "step": 1334
    },
    {
      "completion_length": 100.15625,
      "epoch": 0.8607350096711799,
      "grad_norm": 19.142066955566406,
      "kl": 0.12548828125,
      "learning_rate": 5.6963249516441e-07,
      "loss": 0.005,
      "reward": 1.6992515325546265,
      "reward_std": 0.059500470757484436,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6992515027523041,
      "step": 1335
    },
    {
      "completion_length": 100.671875,
      "epoch": 0.8613797549967763,
      "grad_norm": 9.580693244934082,
      "kl": 0.1220703125,
      "learning_rate": 5.693101225016118e-07,
      "loss": 0.0049,
      "reward": 1.485615611076355,
      "reward_std": 0.08089196868240833,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48561564087867737,
      "step": 1336
    },
    {
      "completion_length": 100.78125,
      "epoch": 0.8620245003223727,
      "grad_norm": 23.950700759887695,
      "kl": 0.12158203125,
      "learning_rate": 5.689877498388136e-07,
      "loss": 0.0049,
      "reward": 1.5655575394630432,
      "reward_std": 0.10747053101658821,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5655575394630432,
      "step": 1337
    },
    {
      "completion_length": 103.96875,
      "epoch": 0.8626692456479691,
      "grad_norm": 21.37566566467285,
      "kl": 0.11181640625,
      "learning_rate": 5.686653771760155e-07,
      "loss": 0.0045,
      "reward": 1.5508422255516052,
      "reward_std": 0.12930458039045334,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5664671659469604,
      "step": 1338
    },
    {
      "completion_length": 102.921875,
      "epoch": 0.8633139909735654,
      "grad_norm": 31.438480377197266,
      "kl": 0.1279296875,
      "learning_rate": 5.683430045132172e-07,
      "loss": 0.0051,
      "reward": 1.6030420064926147,
      "reward_std": 0.10726083442568779,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6030420064926147,
      "step": 1339
    },
    {
      "completion_length": 105.125,
      "epoch": 0.8639587362991619,
      "grad_norm": 29.107524871826172,
      "kl": 0.12353515625,
      "learning_rate": 5.680206318504191e-07,
      "loss": 0.005,
      "reward": 1.6503074169158936,
      "reward_std": 0.08372955396771431,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6503073573112488,
      "step": 1340
    },
    {
      "completion_length": 96.734375,
      "epoch": 0.8646034816247582,
      "grad_norm": 7.437849044799805,
      "kl": 0.130859375,
      "learning_rate": 5.676982591876208e-07,
      "loss": 0.0052,
      "reward": 1.7455703020095825,
      "reward_std": 0.10204130783677101,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7455704212188721,
      "step": 1341
    },
    {
      "completion_length": 105.328125,
      "epoch": 0.8652482269503546,
      "grad_norm": 16.067913055419922,
      "kl": 0.11181640625,
      "learning_rate": 5.673758865248227e-07,
      "loss": 0.0045,
      "reward": 1.4903101921081543,
      "reward_std": 0.14681512117385864,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49031007289886475,
      "step": 1342
    },
    {
      "completion_length": 93.828125,
      "epoch": 0.865892972275951,
      "grad_norm": 23.24372100830078,
      "kl": 0.1328125,
      "learning_rate": 5.670535138620244e-07,
      "loss": 0.0053,
      "reward": 1.554599404335022,
      "reward_std": 0.18923930823802948,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5702243745326996,
      "step": 1343
    },
    {
      "completion_length": 109.75,
      "epoch": 0.8665377176015474,
      "grad_norm": 16.106502532958984,
      "kl": 0.110107421875,
      "learning_rate": 5.667311411992263e-07,
      "loss": 0.0044,
      "reward": 1.7196393609046936,
      "reward_std": 0.12335233762860298,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7196394205093384,
      "step": 1344
    },
    {
      "completion_length": 93.890625,
      "epoch": 0.8671824629271437,
      "grad_norm": 20.337154388427734,
      "kl": 0.32861328125,
      "learning_rate": 5.66408768536428e-07,
      "loss": 0.0131,
      "reward": 1.4278218150138855,
      "reward_std": 0.09321388974785805,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4278218150138855,
      "step": 1345
    },
    {
      "completion_length": 93.171875,
      "epoch": 0.8678272082527402,
      "grad_norm": 9.359206199645996,
      "kl": 0.15576171875,
      "learning_rate": 5.6608639587363e-07,
      "loss": 0.0062,
      "reward": 1.7311944365501404,
      "reward_std": 0.1090300865471363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.731194406747818,
      "step": 1346
    },
    {
      "completion_length": 102.703125,
      "epoch": 0.8684719535783365,
      "grad_norm": 14.245612144470215,
      "kl": 0.12841796875,
      "learning_rate": 5.657640232108317e-07,
      "loss": 0.0051,
      "reward": 1.5649824142456055,
      "reward_std": 0.12245436012744904,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5649823844432831,
      "step": 1347
    },
    {
      "completion_length": 92.203125,
      "epoch": 0.869116698903933,
      "grad_norm": 38.64447784423828,
      "kl": 0.1328125,
      "learning_rate": 5.654416505480335e-07,
      "loss": 0.0053,
      "reward": 1.5066606998443604,
      "reward_std": 0.14087754115462303,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5222857445478439,
      "step": 1348
    },
    {
      "completion_length": 102.828125,
      "epoch": 0.8697614442295294,
      "grad_norm": 15.142080307006836,
      "kl": 0.154052734375,
      "learning_rate": 5.651192778852352e-07,
      "loss": 0.0062,
      "reward": 1.391581654548645,
      "reward_std": 0.1948801502585411,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4072067067027092,
      "step": 1349
    },
    {
      "completion_length": 101.875,
      "epoch": 0.8704061895551257,
      "grad_norm": 8.289034843444824,
      "kl": 0.119140625,
      "learning_rate": 5.647969052224371e-07,
      "loss": 0.0048,
      "reward": 1.6482349634170532,
      "reward_std": 0.11315581947565079,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6482349336147308,
      "step": 1350
    },
    {
      "completion_length": 104.28125,
      "epoch": 0.8710509348807222,
      "grad_norm": 15.9578857421875,
      "kl": 0.11279296875,
      "learning_rate": 5.644745325596389e-07,
      "loss": 0.0045,
      "reward": 1.6309210658073425,
      "reward_std": 0.1061040423810482,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6309210360050201,
      "step": 1351
    },
    {
      "completion_length": 109.65625,
      "epoch": 0.8716956802063185,
      "grad_norm": 13.198500633239746,
      "kl": 0.1337890625,
      "learning_rate": 5.641521598968407e-07,
      "loss": 0.0054,
      "reward": 1.7125477194786072,
      "reward_std": 0.11557654291391373,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7125477194786072,
      "step": 1352
    },
    {
      "completion_length": 92.40625,
      "epoch": 0.8723404255319149,
      "grad_norm": 14.851712226867676,
      "kl": 0.12646484375,
      "learning_rate": 5.638297872340425e-07,
      "loss": 0.005,
      "reward": 1.6003835201263428,
      "reward_std": 0.12276577949523926,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6160085499286652,
      "step": 1353
    },
    {
      "completion_length": 116.03125,
      "epoch": 0.8729851708575113,
      "grad_norm": 15.63801383972168,
      "kl": 0.115478515625,
      "learning_rate": 5.635074145712443e-07,
      "loss": 0.0046,
      "reward": 1.5839186310768127,
      "reward_std": 0.13456114381551743,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5995436608791351,
      "step": 1354
    },
    {
      "completion_length": 98.453125,
      "epoch": 0.8736299161831077,
      "grad_norm": 19.315853118896484,
      "kl": 0.125732421875,
      "learning_rate": 5.631850419084461e-07,
      "loss": 0.005,
      "reward": 1.6114773154258728,
      "reward_std": 0.09730841219425201,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6114773452281952,
      "step": 1355
    },
    {
      "completion_length": 98.171875,
      "epoch": 0.874274661508704,
      "grad_norm": 14.855340003967285,
      "kl": 0.127197265625,
      "learning_rate": 5.62862669245648e-07,
      "loss": 0.0051,
      "reward": 1.5679059624671936,
      "reward_std": 0.0623544380068779,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5679059624671936,
      "step": 1356
    },
    {
      "completion_length": 101.5,
      "epoch": 0.8749194068343005,
      "grad_norm": 30.180572509765625,
      "kl": 0.11962890625,
      "learning_rate": 5.625402965828497e-07,
      "loss": 0.0048,
      "reward": 1.5438405275344849,
      "reward_std": 0.08588853105902672,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5438405424356461,
      "step": 1357
    },
    {
      "completion_length": 95.125,
      "epoch": 0.8755641521598968,
      "grad_norm": 20.203781127929688,
      "kl": 0.17138671875,
      "learning_rate": 5.622179239200515e-07,
      "loss": 0.0069,
      "reward": 1.580243706703186,
      "reward_std": 0.19309964030981064,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.595868706703186,
      "step": 1358
    },
    {
      "completion_length": 98.0625,
      "epoch": 0.8762088974854932,
      "grad_norm": 7.41533899307251,
      "kl": 0.150390625,
      "learning_rate": 5.618955512572534e-07,
      "loss": 0.006,
      "reward": 1.5056850910186768,
      "reward_std": 0.054740700870752335,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5056850016117096,
      "step": 1359
    },
    {
      "completion_length": 105.328125,
      "epoch": 0.8768536428110896,
      "grad_norm": 24.1795597076416,
      "kl": 0.129638671875,
      "learning_rate": 5.615731785944552e-07,
      "loss": 0.0052,
      "reward": 1.5859038829803467,
      "reward_std": 0.08588533103466034,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5859038531780243,
      "step": 1360
    },
    {
      "completion_length": 95.453125,
      "epoch": 0.877498388136686,
      "grad_norm": 13.868168830871582,
      "kl": 0.1279296875,
      "learning_rate": 5.612508059316569e-07,
      "loss": 0.0051,
      "reward": 1.419808030128479,
      "reward_std": 0.06623131036758423,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4198080450296402,
      "step": 1361
    },
    {
      "completion_length": 91.734375,
      "epoch": 0.8781431334622823,
      "grad_norm": 5.384434700012207,
      "kl": 0.129150390625,
      "learning_rate": 5.609284332688588e-07,
      "loss": 0.0052,
      "reward": 1.4930008053779602,
      "reward_std": 0.12473198212683201,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49300073087215424,
      "step": 1362
    },
    {
      "completion_length": 107.765625,
      "epoch": 0.8787878787878788,
      "grad_norm": 19.84931182861328,
      "kl": 0.19189453125,
      "learning_rate": 5.606060606060605e-07,
      "loss": 0.0077,
      "reward": 1.5273191928863525,
      "reward_std": 0.15590080618858337,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5273192077875137,
      "step": 1363
    },
    {
      "completion_length": 92.78125,
      "epoch": 0.8794326241134752,
      "grad_norm": 22.012100219726562,
      "kl": 0.109130859375,
      "learning_rate": 5.602836879432624e-07,
      "loss": 0.0044,
      "reward": 1.6968834400177002,
      "reward_std": 0.11367316171526909,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6968834698200226,
      "step": 1364
    },
    {
      "completion_length": 95.640625,
      "epoch": 0.8800773694390716,
      "grad_norm": 26.02556800842285,
      "kl": 0.117431640625,
      "learning_rate": 5.599613152804641e-07,
      "loss": 0.0047,
      "reward": 1.464781105518341,
      "reward_std": 0.08185772225260735,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46478110551834106,
      "step": 1365
    },
    {
      "completion_length": 105.15625,
      "epoch": 0.880722114764668,
      "grad_norm": 9.117992401123047,
      "kl": 0.11669921875,
      "learning_rate": 5.59638942617666e-07,
      "loss": 0.0047,
      "reward": 1.732893705368042,
      "reward_std": 0.11800439655780792,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7328937351703644,
      "step": 1366
    },
    {
      "completion_length": 104.15625,
      "epoch": 0.8813668600902643,
      "grad_norm": 17.9821720123291,
      "kl": 0.11962890625,
      "learning_rate": 5.593165699548677e-07,
      "loss": 0.0048,
      "reward": 1.4764457941055298,
      "reward_std": 0.13345470279455185,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47644585371017456,
      "step": 1367
    },
    {
      "completion_length": 99.046875,
      "epoch": 0.8820116054158608,
      "grad_norm": 35.48114013671875,
      "kl": 0.124755859375,
      "learning_rate": 5.589941972920696e-07,
      "loss": 0.005,
      "reward": 1.714375615119934,
      "reward_std": 0.1126624196767807,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7143756151199341,
      "step": 1368
    },
    {
      "completion_length": 92.46875,
      "epoch": 0.8826563507414571,
      "grad_norm": 19.23027992248535,
      "kl": 0.12744140625,
      "learning_rate": 5.586718246292714e-07,
      "loss": 0.0051,
      "reward": 1.53688383102417,
      "reward_std": 0.16378530487418175,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5525089502334595,
      "step": 1369
    },
    {
      "completion_length": 107.203125,
      "epoch": 0.8833010960670535,
      "grad_norm": 18.839021682739258,
      "kl": 0.11083984375,
      "learning_rate": 5.583494519664732e-07,
      "loss": 0.0044,
      "reward": 1.618294894695282,
      "reward_std": 0.10168775171041489,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6182949244976044,
      "step": 1370
    },
    {
      "completion_length": 109.796875,
      "epoch": 0.8839458413926499,
      "grad_norm": 14.45029354095459,
      "kl": 0.119140625,
      "learning_rate": 5.580270793036749e-07,
      "loss": 0.0048,
      "reward": 1.5975992679595947,
      "reward_std": 0.16226022690534592,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5975992381572723,
      "step": 1371
    },
    {
      "completion_length": 98.84375,
      "epoch": 0.8845905867182463,
      "grad_norm": 14.76009750366211,
      "kl": 0.100341796875,
      "learning_rate": 5.577047066408769e-07,
      "loss": 0.004,
      "reward": 1.5008023381233215,
      "reward_std": 0.12654516100883484,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5008023977279663,
      "step": 1372
    },
    {
      "completion_length": 101.03125,
      "epoch": 0.8852353320438426,
      "grad_norm": 40.02851867675781,
      "kl": 0.1328125,
      "learning_rate": 5.573823339780786e-07,
      "loss": 0.0053,
      "reward": 1.5849175453186035,
      "reward_std": 0.13229620456695557,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5849175751209259,
      "step": 1373
    },
    {
      "completion_length": 108.234375,
      "epoch": 0.8858800773694391,
      "grad_norm": 12.737630844116211,
      "kl": 0.119384765625,
      "learning_rate": 5.570599613152804e-07,
      "loss": 0.0048,
      "reward": 1.6538455486297607,
      "reward_std": 0.09892918542027473,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6538455188274384,
      "step": 1374
    },
    {
      "completion_length": 106.4375,
      "epoch": 0.8865248226950354,
      "grad_norm": 31.22701644897461,
      "kl": 0.125,
      "learning_rate": 5.567375886524822e-07,
      "loss": 0.005,
      "reward": 1.4634947776794434,
      "reward_std": 0.09996931068599224,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46349479258060455,
      "step": 1375
    },
    {
      "completion_length": 93.71875,
      "epoch": 0.8871695680206318,
      "grad_norm": 47.132286071777344,
      "kl": 0.125244140625,
      "learning_rate": 5.56415215989684e-07,
      "loss": 0.005,
      "reward": 1.585554599761963,
      "reward_std": 0.06121959537267685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5855547189712524,
      "step": 1376
    },
    {
      "completion_length": 90.203125,
      "epoch": 0.8878143133462283,
      "grad_norm": 16.654361724853516,
      "kl": 0.123046875,
      "learning_rate": 5.560928433268859e-07,
      "loss": 0.0049,
      "reward": 1.5993537306785583,
      "reward_std": 0.07232223823666573,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5993536710739136,
      "step": 1377
    },
    {
      "completion_length": 100.9375,
      "epoch": 0.8884590586718246,
      "grad_norm": 16.512577056884766,
      "kl": 0.12158203125,
      "learning_rate": 5.557704706640877e-07,
      "loss": 0.0049,
      "reward": 1.614881694316864,
      "reward_std": 0.12154518440365791,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.614881694316864,
      "step": 1378
    },
    {
      "completion_length": 102.265625,
      "epoch": 0.8891038039974211,
      "grad_norm": 9.475802421569824,
      "kl": 0.113525390625,
      "learning_rate": 5.554480980012894e-07,
      "loss": 0.0045,
      "reward": 1.6055333614349365,
      "reward_std": 0.1013088971376419,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6055334508419037,
      "step": 1379
    },
    {
      "completion_length": 91.640625,
      "epoch": 0.8897485493230174,
      "grad_norm": 10.056422233581543,
      "kl": 0.115966796875,
      "learning_rate": 5.551257253384912e-07,
      "loss": 0.0046,
      "reward": 1.6401957869529724,
      "reward_std": 0.07066305167973042,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.64019575715065,
      "step": 1380
    },
    {
      "completion_length": 98.59375,
      "epoch": 0.8903932946486138,
      "grad_norm": 21.696550369262695,
      "kl": 0.120849609375,
      "learning_rate": 5.548033526756932e-07,
      "loss": 0.0048,
      "reward": 1.554986298084259,
      "reward_std": 0.14988679066300392,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5549863278865814,
      "step": 1381
    },
    {
      "completion_length": 95.75,
      "epoch": 0.8910380399742102,
      "grad_norm": 37.21528244018555,
      "kl": 0.13671875,
      "learning_rate": 5.544809800128949e-07,
      "loss": 0.0055,
      "reward": 1.6293746829032898,
      "reward_std": 0.21692323684692383,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.6606246829032898,
      "step": 1382
    },
    {
      "completion_length": 108.703125,
      "epoch": 0.8916827852998066,
      "grad_norm": 172.50978088378906,
      "kl": 0.134765625,
      "learning_rate": 5.541586073500967e-07,
      "loss": 0.0054,
      "reward": 1.4939188361167908,
      "reward_std": 0.14137040078639984,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49391883611679077,
      "step": 1383
    },
    {
      "completion_length": 104.546875,
      "epoch": 0.8923275306254029,
      "grad_norm": 110.67353057861328,
      "kl": 0.1123046875,
      "learning_rate": 5.538362346872985e-07,
      "loss": 0.0045,
      "reward": 1.5028228163719177,
      "reward_std": 0.18492045998573303,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.502822756767273,
      "step": 1384
    },
    {
      "completion_length": 97.4375,
      "epoch": 0.8929722759509994,
      "grad_norm": 128.3675079345703,
      "kl": 0.129150390625,
      "learning_rate": 5.535138620245004e-07,
      "loss": 0.0052,
      "reward": 1.7198100686073303,
      "reward_std": 0.13068809360265732,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7198100984096527,
      "step": 1385
    },
    {
      "completion_length": 92.015625,
      "epoch": 0.8936170212765957,
      "grad_norm": 22.55219268798828,
      "kl": 0.13623046875,
      "learning_rate": 5.531914893617021e-07,
      "loss": 0.0055,
      "reward": 1.5806416273117065,
      "reward_std": 0.07453634962439537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5806415677070618,
      "step": 1386
    },
    {
      "completion_length": 98.21875,
      "epoch": 0.8942617666021921,
      "grad_norm": 25.168912887573242,
      "kl": 0.109619140625,
      "learning_rate": 5.52869116698904e-07,
      "loss": 0.0044,
      "reward": 1.4894877076148987,
      "reward_std": 0.11500835418701172,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48948773741722107,
      "step": 1387
    },
    {
      "completion_length": 111.9375,
      "epoch": 0.8949065119277885,
      "grad_norm": 8.519505500793457,
      "kl": 0.12353515625,
      "learning_rate": 5.525467440361057e-07,
      "loss": 0.0049,
      "reward": 1.4606132507324219,
      "reward_std": 0.10788729041814804,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.46061332523822784,
      "step": 1388
    },
    {
      "completion_length": 102.515625,
      "epoch": 0.8955512572533849,
      "grad_norm": 25.281314849853516,
      "kl": 0.13330078125,
      "learning_rate": 5.522243713733076e-07,
      "loss": 0.0053,
      "reward": 1.59192955493927,
      "reward_std": 0.14972123503684998,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5919295251369476,
      "step": 1389
    },
    {
      "completion_length": 92.4375,
      "epoch": 0.8961960025789814,
      "grad_norm": 18.566326141357422,
      "kl": 0.22119140625,
      "learning_rate": 5.519019987105093e-07,
      "loss": 0.0088,
      "reward": 1.6087493300437927,
      "reward_std": 0.16570261120796204,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6243743300437927,
      "step": 1390
    },
    {
      "completion_length": 101.21875,
      "epoch": 0.8968407479045777,
      "grad_norm": 37.18821716308594,
      "kl": 0.12841796875,
      "learning_rate": 5.515796260477112e-07,
      "loss": 0.0051,
      "reward": 1.4420801401138306,
      "reward_std": 0.12759844586253166,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44208016991615295,
      "step": 1391
    },
    {
      "completion_length": 96.515625,
      "epoch": 0.8974854932301741,
      "grad_norm": 57.861751556396484,
      "kl": 0.12353515625,
      "learning_rate": 5.512572533849129e-07,
      "loss": 0.0049,
      "reward": 1.4791641235351562,
      "reward_std": 0.05819516442716122,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4791640341281891,
      "step": 1392
    },
    {
      "completion_length": 99.4375,
      "epoch": 0.8981302385557705,
      "grad_norm": 11.22120189666748,
      "kl": 0.13427734375,
      "learning_rate": 5.509348807221147e-07,
      "loss": 0.0054,
      "reward": 1.5360617637634277,
      "reward_std": 0.0876328144222498,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5360617637634277,
      "step": 1393
    },
    {
      "completion_length": 99.46875,
      "epoch": 0.8987749838813669,
      "grad_norm": 9.769732475280762,
      "kl": 0.14111328125,
      "learning_rate": 5.506125080593166e-07,
      "loss": 0.0056,
      "reward": 1.6172817945480347,
      "reward_std": 0.13143368437886238,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6172817945480347,
      "step": 1394
    },
    {
      "completion_length": 96.046875,
      "epoch": 0.8994197292069632,
      "grad_norm": 21.860910415649414,
      "kl": 0.110107421875,
      "learning_rate": 5.502901353965184e-07,
      "loss": 0.0044,
      "reward": 1.7212462425231934,
      "reward_std": 0.09481841512024403,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7212462425231934,
      "step": 1395
    },
    {
      "completion_length": 96.359375,
      "epoch": 0.9000644745325597,
      "grad_norm": 15.907546043395996,
      "kl": 0.121826171875,
      "learning_rate": 5.499677627337201e-07,
      "loss": 0.0049,
      "reward": 1.694203495979309,
      "reward_std": 0.09414593875408173,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6942034661769867,
      "step": 1396
    },
    {
      "completion_length": 107.34375,
      "epoch": 0.900709219858156,
      "grad_norm": 12.596553802490234,
      "kl": 0.119140625,
      "learning_rate": 5.49645390070922e-07,
      "loss": 0.0048,
      "reward": 1.4774943590164185,
      "reward_std": 0.08623380959033966,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4774942696094513,
      "step": 1397
    },
    {
      "completion_length": 89.625,
      "epoch": 0.9013539651837524,
      "grad_norm": 16.927295684814453,
      "kl": 0.13037109375,
      "learning_rate": 5.493230174081238e-07,
      "loss": 0.0052,
      "reward": 1.6467631459236145,
      "reward_std": 0.09450025856494904,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6467630863189697,
      "step": 1398
    },
    {
      "completion_length": 101.609375,
      "epoch": 0.9019987105093488,
      "grad_norm": 14.867881774902344,
      "kl": 0.125,
      "learning_rate": 5.490006447453256e-07,
      "loss": 0.005,
      "reward": 1.5352025032043457,
      "reward_std": 0.08666136488318443,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5352024734020233,
      "step": 1399
    },
    {
      "completion_length": 97.96875,
      "epoch": 0.9026434558349452,
      "grad_norm": 79.15862274169922,
      "kl": 0.13330078125,
      "learning_rate": 5.486782720825274e-07,
      "loss": 0.0053,
      "reward": 1.525882363319397,
      "reward_std": 0.08245071768760681,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5258822739124298,
      "step": 1400
    },
    {
      "completion_length": 90.578125,
      "epoch": 0.9032882011605415,
      "grad_norm": 53.31522750854492,
      "kl": 0.1435546875,
      "learning_rate": 5.483558994197292e-07,
      "loss": 0.0057,
      "reward": 1.5624061822891235,
      "reward_std": 0.13468413054943085,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5624062120914459,
      "step": 1401
    },
    {
      "completion_length": 95.75,
      "epoch": 0.903932946486138,
      "grad_norm": 11.083121299743652,
      "kl": 0.11669921875,
      "learning_rate": 5.48033526756931e-07,
      "loss": 0.0047,
      "reward": 1.6812357902526855,
      "reward_std": 0.13949067145586014,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.681235671043396,
      "step": 1402
    },
    {
      "completion_length": 86.953125,
      "epoch": 0.9045776918117344,
      "grad_norm": 16.63957405090332,
      "kl": 0.1337890625,
      "learning_rate": 5.477111540941329e-07,
      "loss": 0.0053,
      "reward": 1.7723530530929565,
      "reward_std": 0.06749296560883522,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7723531126976013,
      "step": 1403
    },
    {
      "completion_length": 95.046875,
      "epoch": 0.9052224371373307,
      "grad_norm": 23.644563674926758,
      "kl": 0.134521484375,
      "learning_rate": 5.473887814313346e-07,
      "loss": 0.0054,
      "reward": 1.4659785032272339,
      "reward_std": 0.1834467351436615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4659785330295563,
      "step": 1404
    },
    {
      "completion_length": 97.796875,
      "epoch": 0.9058671824629272,
      "grad_norm": 9.021123886108398,
      "kl": 0.11865234375,
      "learning_rate": 5.470664087685364e-07,
      "loss": 0.0048,
      "reward": 1.6176230907440186,
      "reward_std": 0.13125622272491455,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6176231205463409,
      "step": 1405
    },
    {
      "completion_length": 103.875,
      "epoch": 0.9065119277885235,
      "grad_norm": 22.73553466796875,
      "kl": 0.12451171875,
      "learning_rate": 5.467440361057382e-07,
      "loss": 0.005,
      "reward": 1.605991005897522,
      "reward_std": 0.09940499439835548,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.605991005897522,
      "step": 1406
    },
    {
      "completion_length": 85.140625,
      "epoch": 0.90715667311412,
      "grad_norm": 11.063023567199707,
      "kl": 0.135009765625,
      "learning_rate": 5.464216634429401e-07,
      "loss": 0.0054,
      "reward": 1.6347458958625793,
      "reward_std": 0.08830403909087181,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6347458362579346,
      "step": 1407
    },
    {
      "completion_length": 97.765625,
      "epoch": 0.9078014184397163,
      "grad_norm": 22.741058349609375,
      "kl": 0.15283203125,
      "learning_rate": 5.460992907801418e-07,
      "loss": 0.0061,
      "reward": 1.6160573959350586,
      "reward_std": 0.10062050446867943,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6160573959350586,
      "step": 1408
    },
    {
      "completion_length": 98.625,
      "epoch": 0.9084461637653127,
      "grad_norm": 26.559572219848633,
      "kl": 0.1376953125,
      "learning_rate": 5.457769181173437e-07,
      "loss": 0.0055,
      "reward": 1.7845659255981445,
      "reward_std": 0.07418366987258196,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7845659852027893,
      "step": 1409
    },
    {
      "completion_length": 97.9375,
      "epoch": 0.9090909090909091,
      "grad_norm": 9.306669235229492,
      "kl": 0.1171875,
      "learning_rate": 5.454545454545454e-07,
      "loss": 0.0047,
      "reward": 1.5508063435554504,
      "reward_std": 0.09839482605457306,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5508063286542892,
      "step": 1410
    },
    {
      "completion_length": 96.078125,
      "epoch": 0.9097356544165055,
      "grad_norm": 17.91299057006836,
      "kl": 0.177734375,
      "learning_rate": 5.451321727917473e-07,
      "loss": 0.0071,
      "reward": 1.5460429787635803,
      "reward_std": 0.1461724489927292,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5460430383682251,
      "step": 1411
    },
    {
      "completion_length": 99.3125,
      "epoch": 0.9103803997421018,
      "grad_norm": 11.548925399780273,
      "kl": 0.12109375,
      "learning_rate": 5.44809800128949e-07,
      "loss": 0.0048,
      "reward": 1.5693344473838806,
      "reward_std": 0.1188022643327713,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5849594324827194,
      "step": 1412
    },
    {
      "completion_length": 108.453125,
      "epoch": 0.9110251450676983,
      "grad_norm": 11.278193473815918,
      "kl": 0.11572265625,
      "learning_rate": 5.444874274661509e-07,
      "loss": 0.0046,
      "reward": 1.5335027575492859,
      "reward_std": 0.0699017271399498,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5335027575492859,
      "step": 1413
    },
    {
      "completion_length": 105.0625,
      "epoch": 0.9116698903932946,
      "grad_norm": 11.437976837158203,
      "kl": 0.131591796875,
      "learning_rate": 5.441650548033526e-07,
      "loss": 0.0053,
      "reward": 1.6035997867584229,
      "reward_std": 0.07448092475533485,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6035997867584229,
      "step": 1414
    },
    {
      "completion_length": 106.171875,
      "epoch": 0.912314635718891,
      "grad_norm": 21.975309371948242,
      "kl": 0.121826171875,
      "learning_rate": 5.438426821405545e-07,
      "loss": 0.0049,
      "reward": 1.6646922826766968,
      "reward_std": 0.11215503513813019,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6646923124790192,
      "step": 1415
    },
    {
      "completion_length": 104.234375,
      "epoch": 0.9129593810444874,
      "grad_norm": 11.85417366027832,
      "kl": 0.1279296875,
      "learning_rate": 5.435203094777563e-07,
      "loss": 0.0051,
      "reward": 1.6083498001098633,
      "reward_std": 0.06887305155396461,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6083498597145081,
      "step": 1416
    },
    {
      "completion_length": 106.953125,
      "epoch": 0.9136041263700838,
      "grad_norm": 59.94190979003906,
      "kl": 0.12646484375,
      "learning_rate": 5.431979368149581e-07,
      "loss": 0.0051,
      "reward": 1.5003491640090942,
      "reward_std": 0.15336010605096817,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5003491938114166,
      "step": 1417
    },
    {
      "completion_length": 108.390625,
      "epoch": 0.9142488716956803,
      "grad_norm": 10.740958213806152,
      "kl": 0.11181640625,
      "learning_rate": 5.428755641521598e-07,
      "loss": 0.0045,
      "reward": 1.6392948627471924,
      "reward_std": 0.12374603003263474,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.65491983294487,
      "step": 1418
    },
    {
      "completion_length": 96.546875,
      "epoch": 0.9148936170212766,
      "grad_norm": 10.774832725524902,
      "kl": 0.124267578125,
      "learning_rate": 5.425531914893617e-07,
      "loss": 0.005,
      "reward": 1.487897515296936,
      "reward_std": 0.1082611046731472,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48789748549461365,
      "step": 1419
    },
    {
      "completion_length": 94.03125,
      "epoch": 0.915538362346873,
      "grad_norm": 17.441375732421875,
      "kl": 0.24755859375,
      "learning_rate": 5.422308188265635e-07,
      "loss": 0.0099,
      "reward": 1.5370892882347107,
      "reward_std": 0.13201788812875748,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5370892584323883,
      "step": 1420
    },
    {
      "completion_length": 102.203125,
      "epoch": 0.9161831076724694,
      "grad_norm": 23.82035255432129,
      "kl": 0.12451171875,
      "learning_rate": 5.419084461637653e-07,
      "loss": 0.005,
      "reward": 1.5023330450057983,
      "reward_std": 0.08557785674929619,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5023330599069595,
      "step": 1421
    },
    {
      "completion_length": 104.8125,
      "epoch": 0.9168278529980658,
      "grad_norm": 11.59601879119873,
      "kl": 0.122802734375,
      "learning_rate": 5.415860735009671e-07,
      "loss": 0.0049,
      "reward": 1.7010485529899597,
      "reward_std": 0.07506170868873596,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7010485231876373,
      "step": 1422
    },
    {
      "completion_length": 98.140625,
      "epoch": 0.9174725983236621,
      "grad_norm": 21.762170791625977,
      "kl": 0.11767578125,
      "learning_rate": 5.412637008381689e-07,
      "loss": 0.0047,
      "reward": 1.479046642780304,
      "reward_std": 0.09455282241106033,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47904662787914276,
      "step": 1423
    },
    {
      "completion_length": 103.890625,
      "epoch": 0.9181173436492586,
      "grad_norm": 12.237387657165527,
      "kl": 0.1318359375,
      "learning_rate": 5.409413281753707e-07,
      "loss": 0.0053,
      "reward": 1.6916875839233398,
      "reward_std": 0.0679397452622652,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6916875839233398,
      "step": 1424
    },
    {
      "completion_length": 98.34375,
      "epoch": 0.9187620889748549,
      "grad_norm": 13.88647174835205,
      "kl": 0.1220703125,
      "learning_rate": 5.406189555125726e-07,
      "loss": 0.0049,
      "reward": 1.5719730257987976,
      "reward_std": 0.11926516145467758,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.57197305560112,
      "step": 1425
    },
    {
      "completion_length": 93.390625,
      "epoch": 0.9194068343004513,
      "grad_norm": 15.400588989257812,
      "kl": 0.13232421875,
      "learning_rate": 5.402965828497743e-07,
      "loss": 0.0053,
      "reward": 1.651523232460022,
      "reward_std": 0.12631524726748466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.651523232460022,
      "step": 1426
    },
    {
      "completion_length": 90.34375,
      "epoch": 0.9200515796260477,
      "grad_norm": 32.75628662109375,
      "kl": 0.15234375,
      "learning_rate": 5.399742101869761e-07,
      "loss": 0.0061,
      "reward": 1.6175696849822998,
      "reward_std": 0.13943760097026825,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6331946849822998,
      "step": 1427
    },
    {
      "completion_length": 104.9375,
      "epoch": 0.9206963249516441,
      "grad_norm": 19.23307228088379,
      "kl": 0.132080078125,
      "learning_rate": 5.39651837524178e-07,
      "loss": 0.0053,
      "reward": 1.5814802646636963,
      "reward_std": 0.12079915404319763,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5814802348613739,
      "step": 1428
    },
    {
      "completion_length": 101.484375,
      "epoch": 0.9213410702772404,
      "grad_norm": 13.339950561523438,
      "kl": 0.146484375,
      "learning_rate": 5.393294648613798e-07,
      "loss": 0.0058,
      "reward": 1.6199922561645508,
      "reward_std": 0.10084755718708038,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.619992196559906,
      "step": 1429
    },
    {
      "completion_length": 86.46875,
      "epoch": 0.9219858156028369,
      "grad_norm": 9.657578468322754,
      "kl": 0.13720703125,
      "learning_rate": 5.390070921985815e-07,
      "loss": 0.0055,
      "reward": 1.5761032104492188,
      "reward_std": 0.11388009041547775,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5761031806468964,
      "step": 1430
    },
    {
      "completion_length": 102.34375,
      "epoch": 0.9226305609284333,
      "grad_norm": 77.72435760498047,
      "kl": 0.142822265625,
      "learning_rate": 5.386847195357834e-07,
      "loss": 0.0057,
      "reward": 1.303051471710205,
      "reward_std": 0.08642319589853287,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.30305153876543045,
      "step": 1431
    },
    {
      "completion_length": 106.234375,
      "epoch": 0.9232753062540296,
      "grad_norm": 20.863059997558594,
      "kl": 0.124267578125,
      "learning_rate": 5.383623468729851e-07,
      "loss": 0.005,
      "reward": 1.5418643951416016,
      "reward_std": 0.11117121577262878,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5418643653392792,
      "step": 1432
    },
    {
      "completion_length": 92.125,
      "epoch": 0.9239200515796261,
      "grad_norm": 17.035001754760742,
      "kl": 0.1591796875,
      "learning_rate": 5.38039974210187e-07,
      "loss": 0.0064,
      "reward": 1.6256510615348816,
      "reward_std": 0.15372055768966675,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6256510317325592,
      "step": 1433
    },
    {
      "completion_length": 103.953125,
      "epoch": 0.9245647969052224,
      "grad_norm": 10.168933868408203,
      "kl": 0.1123046875,
      "learning_rate": 5.377176015473887e-07,
      "loss": 0.0045,
      "reward": 1.5418055057525635,
      "reward_std": 0.09734742343425751,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5418055355548859,
      "step": 1434
    },
    {
      "completion_length": 99.3125,
      "epoch": 0.9252095422308189,
      "grad_norm": 10.835103988647461,
      "kl": 0.1318359375,
      "learning_rate": 5.373952288845906e-07,
      "loss": 0.0053,
      "reward": 1.4514469504356384,
      "reward_std": 0.1669703871011734,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45144692063331604,
      "step": 1435
    },
    {
      "completion_length": 99.109375,
      "epoch": 0.9258542875564152,
      "grad_norm": 21.041080474853516,
      "kl": 0.145751953125,
      "learning_rate": 5.370728562217923e-07,
      "loss": 0.0058,
      "reward": 1.4748035073280334,
      "reward_std": 0.06715789437294006,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47480346262454987,
      "step": 1436
    },
    {
      "completion_length": 105.203125,
      "epoch": 0.9264990328820116,
      "grad_norm": 7.949612617492676,
      "kl": 0.1064453125,
      "learning_rate": 5.367504835589942e-07,
      "loss": 0.0043,
      "reward": 1.6560617685317993,
      "reward_std": 0.06005428917706013,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6560617983341217,
      "step": 1437
    },
    {
      "completion_length": 93.6875,
      "epoch": 0.927143778207608,
      "grad_norm": 16.48297882080078,
      "kl": 0.127685546875,
      "learning_rate": 5.36428110896196e-07,
      "loss": 0.0051,
      "reward": 1.73074871301651,
      "reward_std": 0.12964390963315964,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7307488024234772,
      "step": 1438
    },
    {
      "completion_length": 100.25,
      "epoch": 0.9277885235332044,
      "grad_norm": 12.600117683410645,
      "kl": 0.120849609375,
      "learning_rate": 5.361057382333978e-07,
      "loss": 0.0048,
      "reward": 1.488064706325531,
      "reward_std": 0.09356208145618439,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4880646616220474,
      "step": 1439
    },
    {
      "completion_length": 90.109375,
      "epoch": 0.9284332688588007,
      "grad_norm": 30.321741104125977,
      "kl": 0.14501953125,
      "learning_rate": 5.357833655705995e-07,
      "loss": 0.0058,
      "reward": 1.4567986726760864,
      "reward_std": 0.13803285732865334,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45679864287376404,
      "step": 1440
    },
    {
      "completion_length": 106.015625,
      "epoch": 0.9290780141843972,
      "grad_norm": 26.666709899902344,
      "kl": 0.117431640625,
      "learning_rate": 5.354609929078015e-07,
      "loss": 0.0047,
      "reward": 1.671416163444519,
      "reward_std": 0.08873732388019562,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6714162230491638,
      "step": 1441
    },
    {
      "completion_length": 95.609375,
      "epoch": 0.9297227595099935,
      "grad_norm": 8.158126831054688,
      "kl": 0.109619140625,
      "learning_rate": 5.351386202450032e-07,
      "loss": 0.0044,
      "reward": 1.531356930732727,
      "reward_std": 0.12410330772399902,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5313569456338882,
      "step": 1442
    },
    {
      "completion_length": 92.03125,
      "epoch": 0.9303675048355899,
      "grad_norm": 13.258638381958008,
      "kl": 0.14111328125,
      "learning_rate": 5.34816247582205e-07,
      "loss": 0.0057,
      "reward": 1.5752546787261963,
      "reward_std": 0.11212371289730072,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5752545893192291,
      "step": 1443
    },
    {
      "completion_length": 95.796875,
      "epoch": 0.9310122501611864,
      "grad_norm": 9.991436004638672,
      "kl": 0.120849609375,
      "learning_rate": 5.344938749194068e-07,
      "loss": 0.0048,
      "reward": 1.6488659381866455,
      "reward_std": 0.13059785589575768,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6644909381866455,
      "step": 1444
    },
    {
      "completion_length": 109.4375,
      "epoch": 0.9316569954867827,
      "grad_norm": 19.7529239654541,
      "kl": 0.159423828125,
      "learning_rate": 5.341715022566086e-07,
      "loss": 0.0064,
      "reward": 1.4868982434272766,
      "reward_std": 0.17621616274118423,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.486898273229599,
      "step": 1445
    },
    {
      "completion_length": 106.6875,
      "epoch": 0.9323017408123792,
      "grad_norm": 9.851184844970703,
      "kl": 0.118408203125,
      "learning_rate": 5.338491295938104e-07,
      "loss": 0.0047,
      "reward": 1.5718659162521362,
      "reward_std": 0.12131549417972565,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.571865975856781,
      "step": 1446
    },
    {
      "completion_length": 94.875,
      "epoch": 0.9329464861379755,
      "grad_norm": 15.667919158935547,
      "kl": 0.1357421875,
      "learning_rate": 5.335267569310123e-07,
      "loss": 0.0054,
      "reward": 1.6483543515205383,
      "reward_std": 0.13322065770626068,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6639794409275055,
      "step": 1447
    },
    {
      "completion_length": 102.671875,
      "epoch": 0.9335912314635719,
      "grad_norm": 16.409921646118164,
      "kl": 0.1494140625,
      "learning_rate": 5.33204384268214e-07,
      "loss": 0.006,
      "reward": 1.6805023550987244,
      "reward_std": 0.13565493375062943,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.680502325296402,
      "step": 1448
    },
    {
      "completion_length": 103.71875,
      "epoch": 0.9342359767891683,
      "grad_norm": 21.049869537353516,
      "kl": 0.114501953125,
      "learning_rate": 5.328820116054158e-07,
      "loss": 0.0046,
      "reward": 1.5279292464256287,
      "reward_std": 0.11103372648358345,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5279292613267899,
      "step": 1449
    },
    {
      "completion_length": 98.59375,
      "epoch": 0.9348807221147647,
      "grad_norm": 12.341633796691895,
      "kl": 0.1298828125,
      "learning_rate": 5.325596389426177e-07,
      "loss": 0.0052,
      "reward": 1.686635673046112,
      "reward_std": 0.09568401798605919,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6866356432437897,
      "step": 1450
    },
    {
      "completion_length": 111.109375,
      "epoch": 0.935525467440361,
      "grad_norm": 14.195385932922363,
      "kl": 0.10986328125,
      "learning_rate": 5.322372662798195e-07,
      "loss": 0.0044,
      "reward": 1.6304091215133667,
      "reward_std": 0.07046160846948624,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6304091215133667,
      "step": 1451
    },
    {
      "completion_length": 94.78125,
      "epoch": 0.9361702127659575,
      "grad_norm": 10.250699996948242,
      "kl": 0.13037109375,
      "learning_rate": 5.319148936170212e-07,
      "loss": 0.0052,
      "reward": 1.6717593669891357,
      "reward_std": 0.10181083902716637,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6717594265937805,
      "step": 1452
    },
    {
      "completion_length": 119.796875,
      "epoch": 0.9368149580915538,
      "grad_norm": 13.502083778381348,
      "kl": 0.108154296875,
      "learning_rate": 5.315925209542231e-07,
      "loss": 0.0043,
      "reward": 1.5886726379394531,
      "reward_std": 0.09807108342647552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5886726975440979,
      "step": 1453
    },
    {
      "completion_length": 97.546875,
      "epoch": 0.9374597034171502,
      "grad_norm": 103.66673278808594,
      "kl": 0.140625,
      "learning_rate": 5.312701482914249e-07,
      "loss": 0.0056,
      "reward": 1.6207711100578308,
      "reward_std": 0.1276988536119461,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.620771050453186,
      "step": 1454
    },
    {
      "completion_length": 104.0,
      "epoch": 0.9381044487427466,
      "grad_norm": 33.64444351196289,
      "kl": 0.140869140625,
      "learning_rate": 5.309477756286267e-07,
      "loss": 0.0056,
      "reward": 1.3918054699897766,
      "reward_std": 0.149025097489357,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3918054550886154,
      "step": 1455
    },
    {
      "completion_length": 100.65625,
      "epoch": 0.938749194068343,
      "grad_norm": 168.25440979003906,
      "kl": 0.1552734375,
      "learning_rate": 5.306254029658284e-07,
      "loss": 0.0062,
      "reward": 1.6715111136436462,
      "reward_std": 0.10104092955589294,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6715111136436462,
      "step": 1456
    },
    {
      "completion_length": 98.59375,
      "epoch": 0.9393939393939394,
      "grad_norm": 21.297224044799805,
      "kl": 0.123046875,
      "learning_rate": 5.303030303030303e-07,
      "loss": 0.0049,
      "reward": 1.5171695351600647,
      "reward_std": 0.09040552377700806,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5171695351600647,
      "step": 1457
    },
    {
      "completion_length": 106.8125,
      "epoch": 0.9400386847195358,
      "grad_norm": 21.35308837890625,
      "kl": 0.13671875,
      "learning_rate": 5.29980657640232e-07,
      "loss": 0.0055,
      "reward": 1.6648077964782715,
      "reward_std": 0.16886910796165466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6648077964782715,
      "step": 1458
    },
    {
      "completion_length": 113.234375,
      "epoch": 0.9406834300451322,
      "grad_norm": 9.700336456298828,
      "kl": 0.107666015625,
      "learning_rate": 5.29658284977434e-07,
      "loss": 0.0043,
      "reward": 1.5858876705169678,
      "reward_std": 0.08786206692457199,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5858877599239349,
      "step": 1459
    },
    {
      "completion_length": 112.28125,
      "epoch": 0.9413281753707285,
      "grad_norm": 16.349515914916992,
      "kl": 0.131591796875,
      "learning_rate": 5.293359123146357e-07,
      "loss": 0.0053,
      "reward": 1.5591378808021545,
      "reward_std": 0.1534385234117508,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5747629702091217,
      "step": 1460
    },
    {
      "completion_length": 99.0625,
      "epoch": 0.941972920696325,
      "grad_norm": 15.572367668151855,
      "kl": 0.1181640625,
      "learning_rate": 5.290135396518375e-07,
      "loss": 0.0047,
      "reward": 1.561755359172821,
      "reward_std": 0.04094194434583187,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5617553442716599,
      "step": 1461
    },
    {
      "completion_length": 109.59375,
      "epoch": 0.9426176660219213,
      "grad_norm": 6.7918524742126465,
      "kl": 0.121826171875,
      "learning_rate": 5.286911669890392e-07,
      "loss": 0.0049,
      "reward": 1.5657913088798523,
      "reward_std": 0.059127068147063255,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5657913386821747,
      "step": 1462
    },
    {
      "completion_length": 121.484375,
      "epoch": 0.9432624113475178,
      "grad_norm": 12.832409858703613,
      "kl": 0.13818359375,
      "learning_rate": 5.283687943262412e-07,
      "loss": 0.0055,
      "reward": 1.4953020811080933,
      "reward_std": 0.10412278026342392,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49530209600925446,
      "step": 1463
    },
    {
      "completion_length": 101.125,
      "epoch": 0.9439071566731141,
      "grad_norm": 19.80808448791504,
      "kl": 0.1328125,
      "learning_rate": 5.280464216634429e-07,
      "loss": 0.0053,
      "reward": 1.6220327019691467,
      "reward_std": 0.16693954914808273,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6376576721668243,
      "step": 1464
    },
    {
      "completion_length": 99.515625,
      "epoch": 0.9445519019987105,
      "grad_norm": 26.701688766479492,
      "kl": 0.122314453125,
      "learning_rate": 5.277240490006447e-07,
      "loss": 0.0049,
      "reward": 1.6026424169540405,
      "reward_std": 0.12471824884414673,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6026423424482346,
      "step": 1465
    },
    {
      "completion_length": 96.65625,
      "epoch": 0.9451966473243069,
      "grad_norm": 17.756526947021484,
      "kl": 0.142578125,
      "learning_rate": 5.274016763378465e-07,
      "loss": 0.0057,
      "reward": 1.597671627998352,
      "reward_std": 0.1306580062955618,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5976717472076416,
      "step": 1466
    },
    {
      "completion_length": 97.84375,
      "epoch": 0.9458413926499033,
      "grad_norm": 35.332550048828125,
      "kl": 0.1396484375,
      "learning_rate": 5.270793036750484e-07,
      "loss": 0.0056,
      "reward": 1.5732786655426025,
      "reward_std": 0.08914842084050179,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.573278620839119,
      "step": 1467
    },
    {
      "completion_length": 106.5625,
      "epoch": 0.9464861379754996,
      "grad_norm": 10.186761856079102,
      "kl": 0.11083984375,
      "learning_rate": 5.267569310122501e-07,
      "loss": 0.0044,
      "reward": 1.7234991788864136,
      "reward_std": 0.08161475323140621,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7234991490840912,
      "step": 1468
    },
    {
      "completion_length": 104.578125,
      "epoch": 0.9471308833010961,
      "grad_norm": 11.844374656677246,
      "kl": 0.1640625,
      "learning_rate": 5.26434558349452e-07,
      "loss": 0.0066,
      "reward": 1.7355561256408691,
      "reward_std": 0.07350215315818787,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7355562448501587,
      "step": 1469
    },
    {
      "completion_length": 100.3125,
      "epoch": 0.9477756286266924,
      "grad_norm": 14.433293342590332,
      "kl": 0.12646484375,
      "learning_rate": 5.261121856866537e-07,
      "loss": 0.0051,
      "reward": 1.7054158449172974,
      "reward_std": 0.09924307838082314,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7054157853126526,
      "step": 1470
    },
    {
      "completion_length": 104.078125,
      "epoch": 0.9484203739522888,
      "grad_norm": 13.400076866149902,
      "kl": 0.1201171875,
      "learning_rate": 5.257898130238555e-07,
      "loss": 0.0048,
      "reward": 1.6034408211708069,
      "reward_std": 0.09964192658662796,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6034408509731293,
      "step": 1471
    },
    {
      "completion_length": 105.53125,
      "epoch": 0.9490651192778853,
      "grad_norm": 11.762677192687988,
      "kl": 0.127197265625,
      "learning_rate": 5.254674403610574e-07,
      "loss": 0.0051,
      "reward": 1.5714890956878662,
      "reward_std": 0.1405608281493187,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5871140956878662,
      "step": 1472
    },
    {
      "completion_length": 113.65625,
      "epoch": 0.9497098646034816,
      "grad_norm": 8.321244239807129,
      "kl": 0.11962890625,
      "learning_rate": 5.251450676982592e-07,
      "loss": 0.0048,
      "reward": 1.5024237632751465,
      "reward_std": 0.10825404524803162,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5024236738681793,
      "step": 1473
    },
    {
      "completion_length": 109.03125,
      "epoch": 0.950354609929078,
      "grad_norm": 21.794754028320312,
      "kl": 0.1279296875,
      "learning_rate": 5.248226950354609e-07,
      "loss": 0.0051,
      "reward": 1.5815457105636597,
      "reward_std": 0.16428036987781525,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5815458297729492,
      "step": 1474
    },
    {
      "completion_length": 102.28125,
      "epoch": 0.9509993552546744,
      "grad_norm": 15.166093826293945,
      "kl": 0.137939453125,
      "learning_rate": 5.245003223726628e-07,
      "loss": 0.0055,
      "reward": 1.696491301059723,
      "reward_std": 0.08702771738171577,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6964913010597229,
      "step": 1475
    },
    {
      "completion_length": 106.84375,
      "epoch": 0.9516441005802708,
      "grad_norm": 35.8002815246582,
      "kl": 0.12548828125,
      "learning_rate": 5.241779497098646e-07,
      "loss": 0.005,
      "reward": 1.6845852136611938,
      "reward_std": 0.1240113228559494,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6845852732658386,
      "step": 1476
    },
    {
      "completion_length": 107.390625,
      "epoch": 0.9522888459058672,
      "grad_norm": 12.195111274719238,
      "kl": 0.114013671875,
      "learning_rate": 5.238555770470664e-07,
      "loss": 0.0046,
      "reward": 1.5479717254638672,
      "reward_std": 0.06802619993686676,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5479717254638672,
      "step": 1477
    },
    {
      "completion_length": 109.234375,
      "epoch": 0.9529335912314636,
      "grad_norm": 7.733105182647705,
      "kl": 0.158203125,
      "learning_rate": 5.235332043842681e-07,
      "loss": 0.0063,
      "reward": 1.670875072479248,
      "reward_std": 0.08894642814993858,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6708749830722809,
      "step": 1478
    },
    {
      "completion_length": 118.703125,
      "epoch": 0.9535783365570599,
      "grad_norm": 9.842519760131836,
      "kl": 0.11865234375,
      "learning_rate": 5.2321083172147e-07,
      "loss": 0.0047,
      "reward": 1.5258830785751343,
      "reward_std": 0.11915917694568634,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5258831083774567,
      "step": 1479
    },
    {
      "completion_length": 110.640625,
      "epoch": 0.9542230818826564,
      "grad_norm": 15.017685890197754,
      "kl": 0.131103515625,
      "learning_rate": 5.228884590586718e-07,
      "loss": 0.0052,
      "reward": 1.4207090139389038,
      "reward_std": 0.08460971340537071,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.420709028840065,
      "step": 1480
    },
    {
      "completion_length": 101.015625,
      "epoch": 0.9548678272082527,
      "grad_norm": 9.356352806091309,
      "kl": 0.12255859375,
      "learning_rate": 5.225660863958736e-07,
      "loss": 0.0049,
      "reward": 1.4379160404205322,
      "reward_std": 0.055083997547626495,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4379161298274994,
      "step": 1481
    },
    {
      "completion_length": 112.875,
      "epoch": 0.9555125725338491,
      "grad_norm": 21.275924682617188,
      "kl": 0.1875,
      "learning_rate": 5.222437137330754e-07,
      "loss": 0.0075,
      "reward": 1.198258638381958,
      "reward_std": 0.11420409381389618,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.19825860857963562,
      "step": 1482
    },
    {
      "completion_length": 123.0,
      "epoch": 0.9561573178594455,
      "grad_norm": 9.146618843078613,
      "kl": 0.11328125,
      "learning_rate": 5.219213410702772e-07,
      "loss": 0.0045,
      "reward": 1.4889427423477173,
      "reward_std": 0.08368909172713757,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4889426678419113,
      "step": 1483
    },
    {
      "completion_length": 126.734375,
      "epoch": 0.9568020631850419,
      "grad_norm": 14.507353782653809,
      "kl": 0.114013671875,
      "learning_rate": 5.215989684074789e-07,
      "loss": 0.0046,
      "reward": 1.525374710559845,
      "reward_std": 0.08460210636258125,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5253747850656509,
      "step": 1484
    },
    {
      "completion_length": 113.0625,
      "epoch": 0.9574468085106383,
      "grad_norm": 44.216976165771484,
      "kl": 0.15185546875,
      "learning_rate": 5.212765957446809e-07,
      "loss": 0.0061,
      "reward": 1.5072972178459167,
      "reward_std": 0.06448614783585072,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5072971731424332,
      "step": 1485
    },
    {
      "completion_length": 103.25,
      "epoch": 0.9580915538362347,
      "grad_norm": 14.714810371398926,
      "kl": 0.17919921875,
      "learning_rate": 5.209542230818826e-07,
      "loss": 0.0072,
      "reward": 1.5354995727539062,
      "reward_std": 0.16414112597703934,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5511245578527451,
      "step": 1486
    },
    {
      "completion_length": 117.234375,
      "epoch": 0.9587362991618311,
      "grad_norm": 10.766145706176758,
      "kl": 0.15283203125,
      "learning_rate": 5.206318504190844e-07,
      "loss": 0.0061,
      "reward": 1.5310158729553223,
      "reward_std": 0.16506896913051605,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5466408729553223,
      "step": 1487
    },
    {
      "completion_length": 120.0,
      "epoch": 0.9593810444874274,
      "grad_norm": 11.279646873474121,
      "kl": 0.127685546875,
      "learning_rate": 5.203094777562862e-07,
      "loss": 0.0051,
      "reward": 1.7490030527114868,
      "reward_std": 0.0665210410952568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7490030825138092,
      "step": 1488
    },
    {
      "completion_length": 107.875,
      "epoch": 0.9600257898130239,
      "grad_norm": 9.737847328186035,
      "kl": 0.11962890625,
      "learning_rate": 5.199871050934881e-07,
      "loss": 0.0048,
      "reward": 1.6569666862487793,
      "reward_std": 0.04395459406077862,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6569667458534241,
      "step": 1489
    },
    {
      "completion_length": 115.234375,
      "epoch": 0.9606705351386202,
      "grad_norm": 16.854692459106445,
      "kl": 0.110107421875,
      "learning_rate": 5.196647324306898e-07,
      "loss": 0.0044,
      "reward": 1.7448121309280396,
      "reward_std": 0.0847182609140873,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7448122203350067,
      "step": 1490
    },
    {
      "completion_length": 105.59375,
      "epoch": 0.9613152804642167,
      "grad_norm": 7.6493449211120605,
      "kl": 0.15185546875,
      "learning_rate": 5.193423597678917e-07,
      "loss": 0.0061,
      "reward": 1.5667211413383484,
      "reward_std": 0.17857857793569565,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.582346111536026,
      "step": 1491
    },
    {
      "completion_length": 108.21875,
      "epoch": 0.961960025789813,
      "grad_norm": 47.15330123901367,
      "kl": 0.128662109375,
      "learning_rate": 5.190199871050934e-07,
      "loss": 0.0052,
      "reward": 1.6914491057395935,
      "reward_std": 0.1899641454219818,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7070741355419159,
      "step": 1492
    },
    {
      "completion_length": 128.96875,
      "epoch": 0.9626047711154094,
      "grad_norm": 29.482093811035156,
      "kl": 0.1552734375,
      "learning_rate": 5.186976144422953e-07,
      "loss": 0.0062,
      "reward": 1.5123082399368286,
      "reward_std": 0.08841293677687645,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5123083293437958,
      "step": 1493
    },
    {
      "completion_length": 114.671875,
      "epoch": 0.9632495164410058,
      "grad_norm": 14.688892364501953,
      "kl": 0.118896484375,
      "learning_rate": 5.183752417794971e-07,
      "loss": 0.0048,
      "reward": 1.7382131814956665,
      "reward_std": 0.1165858581662178,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7382131516933441,
      "step": 1494
    },
    {
      "completion_length": 128.5,
      "epoch": 0.9638942617666022,
      "grad_norm": 9.904383659362793,
      "kl": 0.122802734375,
      "learning_rate": 5.180528691166989e-07,
      "loss": 0.0049,
      "reward": 1.5775507688522339,
      "reward_std": 0.09859605878591537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5775507986545563,
      "step": 1495
    },
    {
      "completion_length": 114.125,
      "epoch": 0.9645390070921985,
      "grad_norm": 14.388285636901855,
      "kl": 0.14111328125,
      "learning_rate": 5.177304964539006e-07,
      "loss": 0.0056,
      "reward": 1.7104634642601013,
      "reward_std": 0.12022716552019119,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7104635238647461,
      "step": 1496
    },
    {
      "completion_length": 111.015625,
      "epoch": 0.965183752417795,
      "grad_norm": 21.702617645263672,
      "kl": 0.12939453125,
      "learning_rate": 5.174081237911026e-07,
      "loss": 0.0052,
      "reward": 1.6979969143867493,
      "reward_std": 0.152745820581913,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6979968249797821,
      "step": 1497
    },
    {
      "completion_length": 119.84375,
      "epoch": 0.9658284977433914,
      "grad_norm": 7.377781391143799,
      "kl": 0.11865234375,
      "learning_rate": 5.170857511283043e-07,
      "loss": 0.0047,
      "reward": 1.6159717440605164,
      "reward_std": 0.0942818857729435,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6159717440605164,
      "step": 1498
    },
    {
      "completion_length": 119.109375,
      "epoch": 0.9664732430689877,
      "grad_norm": 19.44676399230957,
      "kl": 0.133056640625,
      "learning_rate": 5.167633784655061e-07,
      "loss": 0.0053,
      "reward": 1.70160710811615,
      "reward_std": 0.1373962201178074,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7172320783138275,
      "step": 1499
    },
    {
      "completion_length": 106.296875,
      "epoch": 0.9671179883945842,
      "grad_norm": 13.977012634277344,
      "kl": 0.14306640625,
      "learning_rate": 5.164410058027078e-07,
      "loss": 0.0057,
      "reward": 1.6375981569290161,
      "reward_std": 0.13657869026064873,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6375981867313385,
      "step": 1500
    },
    {
      "completion_length": 117.03125,
      "epoch": 0.9677627337201805,
      "grad_norm": 13.271944046020508,
      "kl": 0.153564453125,
      "learning_rate": 5.161186331399097e-07,
      "loss": 0.0061,
      "reward": 1.5755575895309448,
      "reward_std": 0.11266742460429668,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5755576193332672,
      "step": 1501
    },
    {
      "completion_length": 116.421875,
      "epoch": 0.968407479045777,
      "grad_norm": 19.823734283447266,
      "kl": 0.135498046875,
      "learning_rate": 5.157962604771115e-07,
      "loss": 0.0054,
      "reward": 1.6668625473976135,
      "reward_std": 0.09966631233692169,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6668625175952911,
      "step": 1502
    },
    {
      "completion_length": 140.203125,
      "epoch": 0.9690522243713733,
      "grad_norm": 16.144290924072266,
      "kl": 0.119384765625,
      "learning_rate": 5.154738878143133e-07,
      "loss": 0.0048,
      "reward": 1.7107030749320984,
      "reward_std": 0.08905570209026337,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.710703045129776,
      "step": 1503
    },
    {
      "completion_length": 124.796875,
      "epoch": 0.9696969696969697,
      "grad_norm": 12.73034381866455,
      "kl": 0.11865234375,
      "learning_rate": 5.151515151515151e-07,
      "loss": 0.0047,
      "reward": 1.4738882780075073,
      "reward_std": 0.09788031131029129,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4738882780075073,
      "step": 1504
    },
    {
      "completion_length": 124.515625,
      "epoch": 0.970341715022566,
      "grad_norm": 13.489100456237793,
      "kl": 0.1337890625,
      "learning_rate": 5.148291424887169e-07,
      "loss": 0.0053,
      "reward": 1.5239415764808655,
      "reward_std": 0.07494577765464783,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5239415764808655,
      "step": 1505
    },
    {
      "completion_length": 127.6875,
      "epoch": 0.9709864603481625,
      "grad_norm": 19.454742431640625,
      "kl": 0.11865234375,
      "learning_rate": 5.145067698259187e-07,
      "loss": 0.0048,
      "reward": 1.414416253566742,
      "reward_std": 0.1128045842051506,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41441625356674194,
      "step": 1506
    },
    {
      "completion_length": 124.640625,
      "epoch": 0.9716312056737588,
      "grad_norm": 9.694129943847656,
      "kl": 0.11767578125,
      "learning_rate": 5.141843971631206e-07,
      "loss": 0.0047,
      "reward": 1.482945740222931,
      "reward_std": 0.1253538653254509,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4985707104206085,
      "step": 1507
    },
    {
      "completion_length": 139.828125,
      "epoch": 0.9722759509993553,
      "grad_norm": 52.3492546081543,
      "kl": 0.12548828125,
      "learning_rate": 5.138620245003223e-07,
      "loss": 0.005,
      "reward": 1.6624096035957336,
      "reward_std": 0.13141599297523499,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6624096035957336,
      "step": 1508
    },
    {
      "completion_length": 132.140625,
      "epoch": 0.9729206963249516,
      "grad_norm": 19.669851303100586,
      "kl": 0.10986328125,
      "learning_rate": 5.135396518375241e-07,
      "loss": 0.0044,
      "reward": 1.52760249376297,
      "reward_std": 0.14120056480169296,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5276025533676147,
      "step": 1509
    },
    {
      "completion_length": 133.453125,
      "epoch": 0.973565441650548,
      "grad_norm": 11.811445236206055,
      "kl": 0.101806640625,
      "learning_rate": 5.13217279174726e-07,
      "loss": 0.0041,
      "reward": 1.542509913444519,
      "reward_std": 0.09525493159890175,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5425098836421967,
      "step": 1510
    },
    {
      "completion_length": 130.0625,
      "epoch": 0.9742101869761445,
      "grad_norm": 14.556682586669922,
      "kl": 0.107666015625,
      "learning_rate": 5.128949065119278e-07,
      "loss": 0.0043,
      "reward": 1.622962474822998,
      "reward_std": 0.1552848145365715,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6385875046253204,
      "step": 1511
    },
    {
      "completion_length": 114.84375,
      "epoch": 0.9748549323017408,
      "grad_norm": 9.852642059326172,
      "kl": 0.12158203125,
      "learning_rate": 5.125725338491295e-07,
      "loss": 0.0049,
      "reward": 1.5891509652137756,
      "reward_std": 0.1219281405210495,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6047759354114532,
      "step": 1512
    },
    {
      "completion_length": 141.40625,
      "epoch": 0.9754996776273372,
      "grad_norm": 11.842811584472656,
      "kl": 0.109375,
      "learning_rate": 5.122501611863314e-07,
      "loss": 0.0044,
      "reward": 1.6655017733573914,
      "reward_std": 0.11256514489650726,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6655017733573914,
      "step": 1513
    },
    {
      "completion_length": 118.015625,
      "epoch": 0.9761444229529336,
      "grad_norm": 27.52327537536621,
      "kl": 0.114501953125,
      "learning_rate": 5.119277885235331e-07,
      "loss": 0.0046,
      "reward": 1.5326758027076721,
      "reward_std": 0.1902296021580696,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5326757878065109,
      "step": 1514
    },
    {
      "completion_length": 116.375,
      "epoch": 0.97678916827853,
      "grad_norm": 16.492992401123047,
      "kl": 0.1318359375,
      "learning_rate": 5.11605415860735e-07,
      "loss": 0.0053,
      "reward": 1.4814838767051697,
      "reward_std": 0.12518265470862389,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4814838469028473,
      "step": 1515
    },
    {
      "completion_length": 119.140625,
      "epoch": 0.9774339136041263,
      "grad_norm": 12.996352195739746,
      "kl": 0.1376953125,
      "learning_rate": 5.112830431979368e-07,
      "loss": 0.0055,
      "reward": 1.533171832561493,
      "reward_std": 0.13848894834518433,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5331718623638153,
      "step": 1516
    },
    {
      "completion_length": 122.671875,
      "epoch": 0.9780786589297228,
      "grad_norm": 9.61275863647461,
      "kl": 0.114990234375,
      "learning_rate": 5.109606705351386e-07,
      "loss": 0.0046,
      "reward": 1.5685065388679504,
      "reward_std": 0.08499204367399216,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5685064792633057,
      "step": 1517
    },
    {
      "completion_length": 128.046875,
      "epoch": 0.9787234042553191,
      "grad_norm": 16.09702491760254,
      "kl": 0.12451171875,
      "learning_rate": 5.106382978723403e-07,
      "loss": 0.005,
      "reward": 1.4926301836967468,
      "reward_std": 0.07699324190616608,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4926301836967468,
      "step": 1518
    },
    {
      "completion_length": 127.15625,
      "epoch": 0.9793681495809156,
      "grad_norm": 17.23007583618164,
      "kl": 0.111083984375,
      "learning_rate": 5.103159252095423e-07,
      "loss": 0.0044,
      "reward": 1.5509237051010132,
      "reward_std": 0.12189146503806114,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.550923690199852,
      "step": 1519
    },
    {
      "completion_length": 117.90625,
      "epoch": 0.9800128949065119,
      "grad_norm": 20.01177406311035,
      "kl": 0.12646484375,
      "learning_rate": 5.09993552546744e-07,
      "loss": 0.0051,
      "reward": 1.551382064819336,
      "reward_std": 0.10219447687268257,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5513819754123688,
      "step": 1520
    },
    {
      "completion_length": 116.5,
      "epoch": 0.9806576402321083,
      "grad_norm": 33.8641357421875,
      "kl": 0.13916015625,
      "learning_rate": 5.096711798839458e-07,
      "loss": 0.0056,
      "reward": 1.624679148197174,
      "reward_std": 0.1435355320572853,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6246791183948517,
      "step": 1521
    },
    {
      "completion_length": 112.9375,
      "epoch": 0.9813023855577047,
      "grad_norm": 10.071717262268066,
      "kl": 0.123779296875,
      "learning_rate": 5.093488072211475e-07,
      "loss": 0.0049,
      "reward": 1.633798360824585,
      "reward_std": 0.0972403883934021,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6337983310222626,
      "step": 1522
    },
    {
      "completion_length": 128.484375,
      "epoch": 0.9819471308833011,
      "grad_norm": 9.838187217712402,
      "kl": 0.2041015625,
      "learning_rate": 5.090264345583495e-07,
      "loss": 0.0082,
      "reward": 1.6361613273620605,
      "reward_std": 0.18265075236558914,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6517862677574158,
      "step": 1523
    },
    {
      "completion_length": 108.71875,
      "epoch": 0.9825918762088974,
      "grad_norm": 15.268589973449707,
      "kl": 0.121337890625,
      "learning_rate": 5.087040618955512e-07,
      "loss": 0.0049,
      "reward": 1.5366319417953491,
      "reward_std": 0.13059881143271923,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5522569417953491,
      "step": 1524
    },
    {
      "completion_length": 120.765625,
      "epoch": 0.9832366215344939,
      "grad_norm": 13.799566268920898,
      "kl": 0.12744140625,
      "learning_rate": 5.08381689232753e-07,
      "loss": 0.0051,
      "reward": 1.5936528444290161,
      "reward_std": 0.16167183220386505,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5936528742313385,
      "step": 1525
    },
    {
      "completion_length": 123.09375,
      "epoch": 0.9838813668600903,
      "grad_norm": 9.85830020904541,
      "kl": 0.169921875,
      "learning_rate": 5.080593165699548e-07,
      "loss": 0.0068,
      "reward": 1.6045432686805725,
      "reward_std": 0.1380324363708496,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6201682984828949,
      "step": 1526
    },
    {
      "completion_length": 119.78125,
      "epoch": 0.9845261121856866,
      "grad_norm": 41.4371452331543,
      "kl": 0.14453125,
      "learning_rate": 5.077369439071566e-07,
      "loss": 0.0058,
      "reward": 1.6698803901672363,
      "reward_std": 0.10253677517175674,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6698803603649139,
      "step": 1527
    },
    {
      "completion_length": 133.8125,
      "epoch": 0.9851708575112831,
      "grad_norm": 8.188080787658691,
      "kl": 0.179443359375,
      "learning_rate": 5.074145712443584e-07,
      "loss": 0.0072,
      "reward": 1.559647560119629,
      "reward_std": 0.09752954542636871,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5596475452184677,
      "step": 1528
    },
    {
      "completion_length": 118.21875,
      "epoch": 0.9858156028368794,
      "grad_norm": 15.319279670715332,
      "kl": 0.115478515625,
      "learning_rate": 5.070921985815603e-07,
      "loss": 0.0046,
      "reward": 1.651024043560028,
      "reward_std": 0.0649954117834568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6510239839553833,
      "step": 1529
    },
    {
      "completion_length": 116.1875,
      "epoch": 0.9864603481624759,
      "grad_norm": 14.552484512329102,
      "kl": 0.1435546875,
      "learning_rate": 5.06769825918762e-07,
      "loss": 0.0057,
      "reward": 1.630611777305603,
      "reward_std": 0.1458933800458908,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6306118071079254,
      "step": 1530
    },
    {
      "completion_length": 109.734375,
      "epoch": 0.9871050934880722,
      "grad_norm": 18.779993057250977,
      "kl": 0.104248046875,
      "learning_rate": 5.064474532559638e-07,
      "loss": 0.0042,
      "reward": 1.6451267004013062,
      "reward_std": 0.06406429223716259,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6451267302036285,
      "step": 1531
    },
    {
      "completion_length": 114.484375,
      "epoch": 0.9877498388136686,
      "grad_norm": 12.041047096252441,
      "kl": 0.11669921875,
      "learning_rate": 5.061250805931657e-07,
      "loss": 0.0047,
      "reward": 1.5772784948349,
      "reward_std": 0.07114759460091591,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5772784948348999,
      "step": 1532
    },
    {
      "completion_length": 99.15625,
      "epoch": 0.988394584139265,
      "grad_norm": 11.344866752624512,
      "kl": 0.1328125,
      "learning_rate": 5.058027079303675e-07,
      "loss": 0.0053,
      "reward": 1.7052232027053833,
      "reward_std": 0.12659567594528198,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7052232325077057,
      "step": 1533
    },
    {
      "completion_length": 128.125,
      "epoch": 0.9890393294648614,
      "grad_norm": 7.520096302032471,
      "kl": 0.111328125,
      "learning_rate": 5.054803352675692e-07,
      "loss": 0.0045,
      "reward": 1.6211850047111511,
      "reward_std": 0.07714429125189781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6211849749088287,
      "step": 1534
    },
    {
      "completion_length": 114.390625,
      "epoch": 0.9896840747904577,
      "grad_norm": 14.49476432800293,
      "kl": 0.10791015625,
      "learning_rate": 5.051579626047711e-07,
      "loss": 0.0043,
      "reward": 1.516048014163971,
      "reward_std": 0.07544976845383644,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5160480439662933,
      "step": 1535
    },
    {
      "completion_length": 99.28125,
      "epoch": 0.9903288201160542,
      "grad_norm": 16.49233627319336,
      "kl": 0.1298828125,
      "learning_rate": 5.048355899419729e-07,
      "loss": 0.0052,
      "reward": 1.582204818725586,
      "reward_std": 0.08421331271529198,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5822047889232635,
      "step": 1536
    },
    {
      "completion_length": 111.703125,
      "epoch": 0.9909735654416505,
      "grad_norm": 16.815492630004883,
      "kl": 0.1298828125,
      "learning_rate": 5.045132172791747e-07,
      "loss": 0.0052,
      "reward": 1.440306544303894,
      "reward_std": 0.17948133870959282,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.45593155920505524,
      "step": 1537
    },
    {
      "completion_length": 100.984375,
      "epoch": 0.9916183107672469,
      "grad_norm": 22.465801239013672,
      "kl": 0.11767578125,
      "learning_rate": 5.041908446163765e-07,
      "loss": 0.0047,
      "reward": 1.5949344038963318,
      "reward_std": 0.09736866503953934,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5949344038963318,
      "step": 1538
    },
    {
      "completion_length": 108.1875,
      "epoch": 0.9922630560928434,
      "grad_norm": 21.445663452148438,
      "kl": 0.117919921875,
      "learning_rate": 5.038684719535783e-07,
      "loss": 0.0047,
      "reward": 1.6617515087127686,
      "reward_std": 0.1481238529086113,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6617515981197357,
      "step": 1539
    },
    {
      "completion_length": 122.078125,
      "epoch": 0.9929078014184397,
      "grad_norm": 11.333296775817871,
      "kl": 0.105712890625,
      "learning_rate": 5.0354609929078e-07,
      "loss": 0.0042,
      "reward": 1.57760488986969,
      "reward_std": 0.08707677200436592,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5776048898696899,
      "step": 1540
    },
    {
      "completion_length": 99.1875,
      "epoch": 0.9935525467440361,
      "grad_norm": 19.648611068725586,
      "kl": 0.135986328125,
      "learning_rate": 5.03223726627982e-07,
      "loss": 0.0054,
      "reward": 1.6852267384529114,
      "reward_std": 0.14017397910356522,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7008517682552338,
      "step": 1541
    },
    {
      "completion_length": 94.828125,
      "epoch": 0.9941972920696325,
      "grad_norm": 29.060779571533203,
      "kl": 0.115966796875,
      "learning_rate": 5.029013539651837e-07,
      "loss": 0.0046,
      "reward": 1.6200896501541138,
      "reward_std": 0.0910437311977148,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6200897097587585,
      "step": 1542
    },
    {
      "completion_length": 113.28125,
      "epoch": 0.9948420373952289,
      "grad_norm": 10.805424690246582,
      "kl": 0.112060546875,
      "learning_rate": 5.025789813023855e-07,
      "loss": 0.0045,
      "reward": 1.5506914258003235,
      "reward_std": 0.083372563123703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5506914258003235,
      "step": 1543
    },
    {
      "completion_length": 105.953125,
      "epoch": 0.9954867827208252,
      "grad_norm": 46.60286331176758,
      "kl": 0.10888671875,
      "learning_rate": 5.022566086395872e-07,
      "loss": 0.0044,
      "reward": 1.71124666929245,
      "reward_std": 0.07887070626020432,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7112466394901276,
      "step": 1544
    },
    {
      "completion_length": 99.28125,
      "epoch": 0.9961315280464217,
      "grad_norm": 15.590272903442383,
      "kl": 0.13134765625,
      "learning_rate": 5.019342359767892e-07,
      "loss": 0.0052,
      "reward": 1.6452555656433105,
      "reward_std": 0.08676682412624359,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6452555060386658,
      "step": 1545
    },
    {
      "completion_length": 97.859375,
      "epoch": 0.996776273372018,
      "grad_norm": 25.825660705566406,
      "kl": 0.13134765625,
      "learning_rate": 5.016118633139909e-07,
      "loss": 0.0053,
      "reward": 1.582628071308136,
      "reward_std": 0.11850284785032272,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5982531011104584,
      "step": 1546
    },
    {
      "completion_length": 100.21875,
      "epoch": 0.9974210186976145,
      "grad_norm": 12.175593376159668,
      "kl": 0.12060546875,
      "learning_rate": 5.012894906511927e-07,
      "loss": 0.0048,
      "reward": 1.489000380039215,
      "reward_std": 0.10180428251624107,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4890003502368927,
      "step": 1547
    },
    {
      "completion_length": 113.71875,
      "epoch": 0.9980657640232108,
      "grad_norm": 10.186779975891113,
      "kl": 0.12109375,
      "learning_rate": 5.009671179883945e-07,
      "loss": 0.0048,
      "reward": 1.687839686870575,
      "reward_std": 0.0874037854373455,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.687839686870575,
      "step": 1548
    },
    {
      "completion_length": 105.109375,
      "epoch": 0.9987105093488072,
      "grad_norm": 15.987135887145996,
      "kl": 0.12060546875,
      "learning_rate": 5.006447453255964e-07,
      "loss": 0.0048,
      "reward": 1.6305145025253296,
      "reward_std": 0.05935047194361687,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.630514532327652,
      "step": 1549
    },
    {
      "completion_length": 113.21875,
      "epoch": 0.9993552546744036,
      "grad_norm": 18.414926528930664,
      "kl": 0.117919921875,
      "learning_rate": 5.003223726627982e-07,
      "loss": 0.0047,
      "reward": 1.5300384759902954,
      "reward_std": 0.06420031562447548,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5300384759902954,
      "step": 1550
    },
    {
      "completion_length": 98.15625,
      "epoch": 1.0,
      "grad_norm": 21.30752944946289,
      "kl": 0.118408203125,
      "learning_rate": 5e-07,
      "loss": 0.0047,
      "reward": 1.5693151354789734,
      "reward_std": 0.13229862600564957,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5693151652812958,
      "step": 1551
    },
    {
      "completion_length": 125.671875,
      "epoch": 1.0006447453255964,
      "grad_norm": 14.886545181274414,
      "kl": 0.1435546875,
      "learning_rate": 4.996776273372017e-07,
      "loss": 0.0058,
      "reward": 1.620706021785736,
      "reward_std": 0.07777786627411842,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6207059621810913,
      "step": 1552
    },
    {
      "completion_length": 105.84375,
      "epoch": 1.0012894906511929,
      "grad_norm": 31.582460403442383,
      "kl": 0.107421875,
      "learning_rate": 4.993552546744036e-07,
      "loss": 0.0043,
      "reward": 1.5983824729919434,
      "reward_std": 0.10737652331590652,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.598382443189621,
      "step": 1553
    },
    {
      "completion_length": 109.546875,
      "epoch": 1.001934235976789,
      "grad_norm": 19.191818237304688,
      "kl": 0.12451171875,
      "learning_rate": 4.990328820116054e-07,
      "loss": 0.005,
      "reward": 1.5432923436164856,
      "reward_std": 0.14936043322086334,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.543292373418808,
      "step": 1554
    },
    {
      "completion_length": 110.75,
      "epoch": 1.0025789813023855,
      "grad_norm": 8.744702339172363,
      "kl": 0.110595703125,
      "learning_rate": 4.987105093488072e-07,
      "loss": 0.0044,
      "reward": 1.377943515777588,
      "reward_std": 0.046197013929486275,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3779434710741043,
      "step": 1555
    },
    {
      "completion_length": 119.171875,
      "epoch": 1.003223726627982,
      "grad_norm": 18.153520584106445,
      "kl": 0.123046875,
      "learning_rate": 4.98388136686009e-07,
      "loss": 0.0049,
      "reward": 1.532510221004486,
      "reward_std": 0.13577953726053238,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5325101613998413,
      "step": 1556
    },
    {
      "completion_length": 118.078125,
      "epoch": 1.0038684719535784,
      "grad_norm": 5.668532848358154,
      "kl": 0.1162109375,
      "learning_rate": 4.980657640232109e-07,
      "loss": 0.0046,
      "reward": 1.4012870788574219,
      "reward_std": 0.05575377307832241,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4012870639562607,
      "step": 1557
    },
    {
      "completion_length": 118.96875,
      "epoch": 1.0045132172791746,
      "grad_norm": 21.61020851135254,
      "kl": 0.12060546875,
      "learning_rate": 4.977433913604126e-07,
      "loss": 0.0048,
      "reward": 1.5471839904785156,
      "reward_std": 0.12884113937616348,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5471839904785156,
      "step": 1558
    },
    {
      "completion_length": 119.75,
      "epoch": 1.005157962604771,
      "grad_norm": 11.653385162353516,
      "kl": 0.24658203125,
      "learning_rate": 4.974210186976144e-07,
      "loss": 0.0099,
      "reward": 1.6526493430137634,
      "reward_std": 0.05117896664887667,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.652649313211441,
      "step": 1559
    },
    {
      "completion_length": 115.625,
      "epoch": 1.0058027079303675,
      "grad_norm": 14.82691478729248,
      "kl": 0.1259765625,
      "learning_rate": 4.970986460348163e-07,
      "loss": 0.005,
      "reward": 1.6142604351043701,
      "reward_std": 0.10050322860479355,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6142604947090149,
      "step": 1560
    },
    {
      "completion_length": 111.09375,
      "epoch": 1.006447453255964,
      "grad_norm": 8.875225067138672,
      "kl": 0.11376953125,
      "learning_rate": 4.967762733720181e-07,
      "loss": 0.0046,
      "reward": 1.7486295104026794,
      "reward_std": 0.08543649315834045,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7486294507980347,
      "step": 1561
    },
    {
      "completion_length": 123.1875,
      "epoch": 1.0070921985815602,
      "grad_norm": 15.650257110595703,
      "kl": 0.12060546875,
      "learning_rate": 4.964539007092198e-07,
      "loss": 0.0048,
      "reward": 1.564130961894989,
      "reward_std": 0.09400678798556328,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.564130961894989,
      "step": 1562
    },
    {
      "completion_length": 112.328125,
      "epoch": 1.0077369439071566,
      "grad_norm": 17.488706588745117,
      "kl": 0.117431640625,
      "learning_rate": 4.961315280464217e-07,
      "loss": 0.0047,
      "reward": 1.5162153840065002,
      "reward_std": 0.1196032203733921,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5162154138088226,
      "step": 1563
    },
    {
      "completion_length": 110.8125,
      "epoch": 1.008381689232753,
      "grad_norm": 9.31165599822998,
      "kl": 0.11279296875,
      "learning_rate": 4.958091553836235e-07,
      "loss": 0.0045,
      "reward": 1.625089704990387,
      "reward_std": 0.08436309918761253,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6250897943973541,
      "step": 1564
    },
    {
      "completion_length": 118.46875,
      "epoch": 1.0090264345583495,
      "grad_norm": 44.249385833740234,
      "kl": 0.093505859375,
      "learning_rate": 4.954867827208252e-07,
      "loss": 0.0037,
      "reward": 1.625906765460968,
      "reward_std": 0.07646162062883377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6259067952632904,
      "step": 1565
    },
    {
      "completion_length": 110.71875,
      "epoch": 1.009671179883946,
      "grad_norm": 17.5499267578125,
      "kl": 0.134765625,
      "learning_rate": 4.951644100580271e-07,
      "loss": 0.0054,
      "reward": 1.7371680736541748,
      "reward_std": 0.11888208240270615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7371681034564972,
      "step": 1566
    },
    {
      "completion_length": 107.765625,
      "epoch": 1.0103159252095422,
      "grad_norm": 20.713336944580078,
      "kl": 0.13818359375,
      "learning_rate": 4.948420373952289e-07,
      "loss": 0.0055,
      "reward": 1.4837321639060974,
      "reward_std": 0.0846199318766594,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4837321788072586,
      "step": 1567
    },
    {
      "completion_length": 107.328125,
      "epoch": 1.0109606705351386,
      "grad_norm": 23.845182418823242,
      "kl": 0.12060546875,
      "learning_rate": 4.945196647324307e-07,
      "loss": 0.0048,
      "reward": 1.6508268117904663,
      "reward_std": 0.06122574955224991,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6508268117904663,
      "step": 1568
    },
    {
      "completion_length": 125.0625,
      "epoch": 1.011605415860735,
      "grad_norm": 9.649888038635254,
      "kl": 0.10986328125,
      "learning_rate": 4.941972920696324e-07,
      "loss": 0.0044,
      "reward": 1.5454849004745483,
      "reward_std": 0.15033403038978577,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5611098408699036,
      "step": 1569
    },
    {
      "completion_length": 117.015625,
      "epoch": 1.0122501611863315,
      "grad_norm": 14.543222427368164,
      "kl": 0.126220703125,
      "learning_rate": 4.938749194068343e-07,
      "loss": 0.0051,
      "reward": 1.6417489647865295,
      "reward_std": 0.06025063619017601,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6417489647865295,
      "step": 1570
    },
    {
      "completion_length": 104.578125,
      "epoch": 1.0128949065119277,
      "grad_norm": 20.035690307617188,
      "kl": 0.17822265625,
      "learning_rate": 4.935525467440361e-07,
      "loss": 0.0071,
      "reward": 1.7167410850524902,
      "reward_std": 0.13318730890750885,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7167410850524902,
      "step": 1571
    },
    {
      "completion_length": 116.296875,
      "epoch": 1.0135396518375241,
      "grad_norm": 7.308071136474609,
      "kl": 0.13427734375,
      "learning_rate": 4.93230174081238e-07,
      "loss": 0.0054,
      "reward": 1.4862470030784607,
      "reward_std": 0.11065686121582985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4862470328807831,
      "step": 1572
    },
    {
      "completion_length": 107.84375,
      "epoch": 1.0141843971631206,
      "grad_norm": 9.348837852478027,
      "kl": 0.1064453125,
      "learning_rate": 4.929078014184397e-07,
      "loss": 0.0043,
      "reward": 1.4404972791671753,
      "reward_std": 0.08681789599359035,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4404972344636917,
      "step": 1573
    },
    {
      "completion_length": 108.046875,
      "epoch": 1.014829142488717,
      "grad_norm": 40.28298568725586,
      "kl": 0.13916015625,
      "learning_rate": 4.925854287556415e-07,
      "loss": 0.0056,
      "reward": 1.7261015176773071,
      "reward_std": 0.11216892302036285,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7261015474796295,
      "step": 1574
    },
    {
      "completion_length": 111.609375,
      "epoch": 1.0154738878143132,
      "grad_norm": 30.62251853942871,
      "kl": 0.11474609375,
      "learning_rate": 4.922630560928433e-07,
      "loss": 0.0046,
      "reward": 1.5981187224388123,
      "reward_std": 0.1484297215938568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5981186926364899,
      "step": 1575
    },
    {
      "completion_length": 107.96875,
      "epoch": 1.0161186331399097,
      "grad_norm": 14.246084213256836,
      "kl": 0.128662109375,
      "learning_rate": 4.919406834300452e-07,
      "loss": 0.0051,
      "reward": 1.5964959263801575,
      "reward_std": 0.09668882563710213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5964959263801575,
      "step": 1576
    },
    {
      "completion_length": 110.015625,
      "epoch": 1.0167633784655061,
      "grad_norm": 45.870033264160156,
      "kl": 0.10986328125,
      "learning_rate": 4.916183107672469e-07,
      "loss": 0.0044,
      "reward": 1.6973443031311035,
      "reward_std": 0.06511173956096172,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6973443031311035,
      "step": 1577
    },
    {
      "completion_length": 115.78125,
      "epoch": 1.0174081237911026,
      "grad_norm": 23.571115493774414,
      "kl": 0.132568359375,
      "learning_rate": 4.912959381044487e-07,
      "loss": 0.0053,
      "reward": 1.6813059449195862,
      "reward_std": 0.1136469803750515,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6813060343265533,
      "step": 1578
    },
    {
      "completion_length": 99.984375,
      "epoch": 1.018052869116699,
      "grad_norm": 42.18523025512695,
      "kl": 0.12353515625,
      "learning_rate": 4.909735654416506e-07,
      "loss": 0.0049,
      "reward": 1.5764471292495728,
      "reward_std": 0.10761763527989388,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5764471590518951,
      "step": 1579
    },
    {
      "completion_length": 113.234375,
      "epoch": 1.0186976144422952,
      "grad_norm": 37.51557159423828,
      "kl": 0.106201171875,
      "learning_rate": 4.906511927788523e-07,
      "loss": 0.0042,
      "reward": 1.6667637825012207,
      "reward_std": 0.12272565066814423,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6667637825012207,
      "step": 1580
    },
    {
      "completion_length": 104.4375,
      "epoch": 1.0193423597678917,
      "grad_norm": 10.304484367370605,
      "kl": 0.12060546875,
      "learning_rate": 4.903288201160541e-07,
      "loss": 0.0048,
      "reward": 1.5135563611984253,
      "reward_std": 0.08310433477163315,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5135563760995865,
      "step": 1581
    },
    {
      "completion_length": 106.28125,
      "epoch": 1.0199871050934881,
      "grad_norm": 17.609773635864258,
      "kl": 0.1142578125,
      "learning_rate": 4.90006447453256e-07,
      "loss": 0.0046,
      "reward": 1.6618885397911072,
      "reward_std": 0.07358003780245781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6618885397911072,
      "step": 1582
    },
    {
      "completion_length": 95.5,
      "epoch": 1.0206318504190846,
      "grad_norm": 39.74180221557617,
      "kl": 0.4404296875,
      "learning_rate": 4.896840747904578e-07,
      "loss": 0.0176,
      "reward": 1.6940548419952393,
      "reward_std": 0.17156479507684708,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6940547823905945,
      "step": 1583
    },
    {
      "completion_length": 103.96875,
      "epoch": 1.0212765957446808,
      "grad_norm": 20.01495933532715,
      "kl": 0.120361328125,
      "learning_rate": 4.893617021276595e-07,
      "loss": 0.0048,
      "reward": 1.537524700164795,
      "reward_std": 0.08335036598145962,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5375246703624725,
      "step": 1584
    },
    {
      "completion_length": 101.890625,
      "epoch": 1.0219213410702772,
      "grad_norm": 14.507052421569824,
      "kl": 0.099609375,
      "learning_rate": 4.890393294648614e-07,
      "loss": 0.004,
      "reward": 1.5363896489143372,
      "reward_std": 0.060725823044776917,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5363896787166595,
      "step": 1585
    },
    {
      "completion_length": 114.59375,
      "epoch": 1.0225660863958737,
      "grad_norm": 22.21466827392578,
      "kl": 0.0986328125,
      "learning_rate": 4.887169568020632e-07,
      "loss": 0.0039,
      "reward": 1.6946098804473877,
      "reward_std": 0.12630461901426315,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6946098208427429,
      "step": 1586
    },
    {
      "completion_length": 102.25,
      "epoch": 1.02321083172147,
      "grad_norm": 11.273469924926758,
      "kl": 0.1396484375,
      "learning_rate": 4.88394584139265e-07,
      "loss": 0.0056,
      "reward": 1.5050395131111145,
      "reward_std": 0.1273115649819374,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.505039632320404,
      "step": 1587
    },
    {
      "completion_length": 101.15625,
      "epoch": 1.0238555770470663,
      "grad_norm": 9.453971862792969,
      "kl": 0.1181640625,
      "learning_rate": 4.880722114764668e-07,
      "loss": 0.0047,
      "reward": 1.6377010941505432,
      "reward_std": 0.17548616230487823,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6533261239528656,
      "step": 1588
    },
    {
      "completion_length": 109.53125,
      "epoch": 1.0245003223726628,
      "grad_norm": 32.01575469970703,
      "kl": 0.1201171875,
      "learning_rate": 4.877498388136686e-07,
      "loss": 0.0048,
      "reward": 1.6303631067276,
      "reward_std": 0.1175847053527832,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6303631365299225,
      "step": 1589
    },
    {
      "completion_length": 104.15625,
      "epoch": 1.0251450676982592,
      "grad_norm": 10.830766677856445,
      "kl": 0.130126953125,
      "learning_rate": 4.874274661508704e-07,
      "loss": 0.0052,
      "reward": 1.4934070110321045,
      "reward_std": 0.13720114901661873,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4934070557355881,
      "step": 1590
    },
    {
      "completion_length": 98.375,
      "epoch": 1.0257898130238556,
      "grad_norm": 47.61631393432617,
      "kl": 0.154296875,
      "learning_rate": 4.871050934880723e-07,
      "loss": 0.0062,
      "reward": 1.6362435221672058,
      "reward_std": 0.13050130754709244,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6362435519695282,
      "step": 1591
    },
    {
      "completion_length": 104.609375,
      "epoch": 1.026434558349452,
      "grad_norm": 25.48038101196289,
      "kl": 0.12060546875,
      "learning_rate": 4.86782720825274e-07,
      "loss": 0.0048,
      "reward": 1.6710693836212158,
      "reward_std": 0.08275764994323254,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6710693538188934,
      "step": 1592
    },
    {
      "completion_length": 97.640625,
      "epoch": 1.0270793036750483,
      "grad_norm": 148.87600708007812,
      "kl": 0.11767578125,
      "learning_rate": 4.864603481624758e-07,
      "loss": 0.0047,
      "reward": 1.4782089591026306,
      "reward_std": 0.17569049820303917,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.478208988904953,
      "step": 1593
    },
    {
      "completion_length": 104.65625,
      "epoch": 1.0277240490006447,
      "grad_norm": 82.6065444946289,
      "kl": 0.110595703125,
      "learning_rate": 4.861379754996777e-07,
      "loss": 0.0044,
      "reward": 1.5708330869674683,
      "reward_std": 0.09587904810905457,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5708329975605011,
      "step": 1594
    },
    {
      "completion_length": 105.703125,
      "epoch": 1.0283687943262412,
      "grad_norm": 12.176362991333008,
      "kl": 0.108642578125,
      "learning_rate": 4.858156028368794e-07,
      "loss": 0.0043,
      "reward": 1.6616841554641724,
      "reward_std": 0.09205985069274902,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6616840362548828,
      "step": 1595
    },
    {
      "completion_length": 101.5,
      "epoch": 1.0290135396518376,
      "grad_norm": 26.574724197387695,
      "kl": 0.1240234375,
      "learning_rate": 4.854932301740812e-07,
      "loss": 0.005,
      "reward": 1.5319926142692566,
      "reward_std": 0.07157671824097633,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5319925844669342,
      "step": 1596
    },
    {
      "completion_length": 104.03125,
      "epoch": 1.0296582849774338,
      "grad_norm": 27.543840408325195,
      "kl": 0.125,
      "learning_rate": 4.85170857511283e-07,
      "loss": 0.005,
      "reward": 1.3573514819145203,
      "reward_std": 0.10726149939000607,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3573514372110367,
      "step": 1597
    },
    {
      "completion_length": 105.25,
      "epoch": 1.0303030303030303,
      "grad_norm": 52.3618049621582,
      "kl": 0.1220703125,
      "learning_rate": 4.848484848484849e-07,
      "loss": 0.0049,
      "reward": 1.6784070134162903,
      "reward_std": 0.06625668704509735,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6784070134162903,
      "step": 1598
    },
    {
      "completion_length": 101.203125,
      "epoch": 1.0309477756286267,
      "grad_norm": 24.95046043395996,
      "kl": 0.1103515625,
      "learning_rate": 4.845261121856866e-07,
      "loss": 0.0044,
      "reward": 1.5376904606819153,
      "reward_std": 0.07565680518746376,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5376904159784317,
      "step": 1599
    },
    {
      "completion_length": 95.953125,
      "epoch": 1.0315925209542232,
      "grad_norm": 24.625473022460938,
      "kl": 0.123779296875,
      "learning_rate": 4.842037395228884e-07,
      "loss": 0.0049,
      "reward": 1.6368153095245361,
      "reward_std": 0.17632479965686798,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6524402499198914,
      "step": 1600
    },
    {
      "completion_length": 113.0,
      "epoch": 1.0322372662798194,
      "grad_norm": 23.780305862426758,
      "kl": 0.120361328125,
      "learning_rate": 4.838813668600903e-07,
      "loss": 0.0048,
      "reward": 1.611727237701416,
      "reward_std": 0.09991315007209778,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6117272675037384,
      "step": 1601
    },
    {
      "completion_length": 100.484375,
      "epoch": 1.0328820116054158,
      "grad_norm": 20.35822296142578,
      "kl": 0.14794921875,
      "learning_rate": 4.835589941972921e-07,
      "loss": 0.0059,
      "reward": 1.5440884828567505,
      "reward_std": 0.11685510165989399,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5440884828567505,
      "step": 1602
    },
    {
      "completion_length": 102.609375,
      "epoch": 1.0335267569310123,
      "grad_norm": 13.551732063293457,
      "kl": 0.142333984375,
      "learning_rate": 4.832366215344938e-07,
      "loss": 0.0057,
      "reward": 1.6955845952033997,
      "reward_std": 0.09659884497523308,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6955845654010773,
      "step": 1603
    },
    {
      "completion_length": 102.703125,
      "epoch": 1.0341715022566087,
      "grad_norm": 14.002148628234863,
      "kl": 0.1396484375,
      "learning_rate": 4.829142488716957e-07,
      "loss": 0.0056,
      "reward": 1.683996319770813,
      "reward_std": 0.17433501780033112,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6839962899684906,
      "step": 1604
    },
    {
      "completion_length": 110.578125,
      "epoch": 1.0348162475822051,
      "grad_norm": 15.583176612854004,
      "kl": 0.126220703125,
      "learning_rate": 4.825918762088975e-07,
      "loss": 0.005,
      "reward": 1.5995303988456726,
      "reward_std": 0.052622975781559944,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.599530428647995,
      "step": 1605
    },
    {
      "completion_length": 99.984375,
      "epoch": 1.0354609929078014,
      "grad_norm": 15.612828254699707,
      "kl": 0.1435546875,
      "learning_rate": 4.822695035460992e-07,
      "loss": 0.0057,
      "reward": 1.6585041880607605,
      "reward_std": 0.11094475910067558,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6585041582584381,
      "step": 1606
    },
    {
      "completion_length": 101.21875,
      "epoch": 1.0361057382333978,
      "grad_norm": 24.793502807617188,
      "kl": 0.13134765625,
      "learning_rate": 4.819471308833011e-07,
      "loss": 0.0052,
      "reward": 1.536777913570404,
      "reward_std": 0.11932965368032455,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5367779284715652,
      "step": 1607
    },
    {
      "completion_length": 107.5625,
      "epoch": 1.0367504835589942,
      "grad_norm": 11.683782577514648,
      "kl": 0.1220703125,
      "learning_rate": 4.816247582205029e-07,
      "loss": 0.0049,
      "reward": 1.6076741814613342,
      "reward_std": 0.17864874005317688,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.607674166560173,
      "step": 1608
    },
    {
      "completion_length": 95.421875,
      "epoch": 1.0373952288845907,
      "grad_norm": 15.090388298034668,
      "kl": 0.133056640625,
      "learning_rate": 4.813023855577047e-07,
      "loss": 0.0053,
      "reward": 1.5302485227584839,
      "reward_std": 0.14502151682972908,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5302484333515167,
      "step": 1609
    },
    {
      "completion_length": 96.53125,
      "epoch": 1.038039974210187,
      "grad_norm": 12.423053741455078,
      "kl": 0.12158203125,
      "learning_rate": 4.809800128949065e-07,
      "loss": 0.0049,
      "reward": 1.6846227645874023,
      "reward_std": 0.10945509001612663,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6846228241920471,
      "step": 1610
    },
    {
      "completion_length": 105.046875,
      "epoch": 1.0386847195357833,
      "grad_norm": 10.02845573425293,
      "kl": 0.121826171875,
      "learning_rate": 4.806576402321083e-07,
      "loss": 0.0049,
      "reward": 1.727286458015442,
      "reward_std": 0.054582754150033,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7272864580154419,
      "step": 1611
    },
    {
      "completion_length": 100.0625,
      "epoch": 1.0393294648613798,
      "grad_norm": 7.871588230133057,
      "kl": 0.1083984375,
      "learning_rate": 4.803352675693101e-07,
      "loss": 0.0043,
      "reward": 1.6807933449745178,
      "reward_std": 0.03430874086916447,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6807933747768402,
      "step": 1612
    },
    {
      "completion_length": 111.375,
      "epoch": 1.0399742101869762,
      "grad_norm": 10.092997550964355,
      "kl": 0.11083984375,
      "learning_rate": 4.80012894906512e-07,
      "loss": 0.0044,
      "reward": 1.5945587754249573,
      "reward_std": 0.08610989525914192,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6101837754249573,
      "step": 1613
    },
    {
      "completion_length": 104.1875,
      "epoch": 1.0406189555125724,
      "grad_norm": 13.337946891784668,
      "kl": 0.1337890625,
      "learning_rate": 4.796905222437137e-07,
      "loss": 0.0054,
      "reward": 1.5582968592643738,
      "reward_std": 0.09042531624436378,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5582968294620514,
      "step": 1614
    },
    {
      "completion_length": 101.421875,
      "epoch": 1.0412637008381689,
      "grad_norm": 18.421772003173828,
      "kl": 0.13671875,
      "learning_rate": 4.793681495809155e-07,
      "loss": 0.0055,
      "reward": 1.5865644216537476,
      "reward_std": 0.07357398793101311,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5865644216537476,
      "step": 1615
    },
    {
      "completion_length": 99.328125,
      "epoch": 1.0419084461637653,
      "grad_norm": 8.236613273620605,
      "kl": 0.140625,
      "learning_rate": 4.790457769181174e-07,
      "loss": 0.0056,
      "reward": 1.5641028881072998,
      "reward_std": 0.12269283831119537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.564102828502655,
      "step": 1616
    },
    {
      "completion_length": 99.609375,
      "epoch": 1.0425531914893618,
      "grad_norm": 12.832371711730957,
      "kl": 0.1455078125,
      "learning_rate": 4.787234042553192e-07,
      "loss": 0.0058,
      "reward": 1.5059439539909363,
      "reward_std": 0.07279989495873451,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5059439837932587,
      "step": 1617
    },
    {
      "completion_length": 115.75,
      "epoch": 1.0431979368149582,
      "grad_norm": 9.09863567352295,
      "kl": 0.134765625,
      "learning_rate": 4.784010315925209e-07,
      "loss": 0.0054,
      "reward": 1.5284171104431152,
      "reward_std": 0.05645672604441643,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5284171551465988,
      "step": 1618
    },
    {
      "completion_length": 99.515625,
      "epoch": 1.0438426821405544,
      "grad_norm": 14.850272178649902,
      "kl": 0.1474609375,
      "learning_rate": 4.780786589297227e-07,
      "loss": 0.0059,
      "reward": 1.6368718147277832,
      "reward_std": 0.14861361682415009,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6368718147277832,
      "step": 1619
    },
    {
      "completion_length": 100.203125,
      "epoch": 1.0444874274661509,
      "grad_norm": 20.048105239868164,
      "kl": 0.13671875,
      "learning_rate": 4.777562862669246e-07,
      "loss": 0.0055,
      "reward": 1.761348843574524,
      "reward_std": 0.1145159900188446,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7613488435745239,
      "step": 1620
    },
    {
      "completion_length": 102.875,
      "epoch": 1.0451321727917473,
      "grad_norm": 10.379176139831543,
      "kl": 0.124755859375,
      "learning_rate": 4.774339136041263e-07,
      "loss": 0.005,
      "reward": 1.5658888816833496,
      "reward_std": 0.08592114225029945,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5658888667821884,
      "step": 1621
    },
    {
      "completion_length": 105.125,
      "epoch": 1.0457769181173437,
      "grad_norm": 15.628901481628418,
      "kl": 0.12841796875,
      "learning_rate": 4.771115409413281e-07,
      "loss": 0.0051,
      "reward": 1.6197215914726257,
      "reward_std": 0.06863688118755817,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6197216510772705,
      "step": 1622
    },
    {
      "completion_length": 107.625,
      "epoch": 1.04642166344294,
      "grad_norm": 11.287335395812988,
      "kl": 0.11474609375,
      "learning_rate": 4.767891682785299e-07,
      "loss": 0.0046,
      "reward": 1.5502777695655823,
      "reward_std": 0.13397856429219246,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5502777099609375,
      "step": 1623
    },
    {
      "completion_length": 105.296875,
      "epoch": 1.0470664087685364,
      "grad_norm": 52.87205505371094,
      "kl": 0.13427734375,
      "learning_rate": 4.7646679561573175e-07,
      "loss": 0.0054,
      "reward": 1.6524277925491333,
      "reward_std": 0.089669119566679,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6524278521537781,
      "step": 1624
    },
    {
      "completion_length": 105.71875,
      "epoch": 1.0477111540941328,
      "grad_norm": 8.991185188293457,
      "kl": 0.11767578125,
      "learning_rate": 4.7614442295293353e-07,
      "loss": 0.0047,
      "reward": 1.8288054466247559,
      "reward_std": 0.08612105995416641,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.8288054168224335,
      "step": 1625
    },
    {
      "completion_length": 98.5,
      "epoch": 1.0483558994197293,
      "grad_norm": 265.6639709472656,
      "kl": 0.14794921875,
      "learning_rate": 4.7582205029013537e-07,
      "loss": 0.0059,
      "reward": 1.721388816833496,
      "reward_std": 0.11115417629480362,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7213887572288513,
      "step": 1626
    },
    {
      "completion_length": 94.34375,
      "epoch": 1.0490006447453255,
      "grad_norm": 55.32572555541992,
      "kl": 0.150390625,
      "learning_rate": 4.7549967762733715e-07,
      "loss": 0.006,
      "reward": 1.7054846286773682,
      "reward_std": 0.06372135505080223,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7054846584796906,
      "step": 1627
    },
    {
      "completion_length": 107.578125,
      "epoch": 1.049645390070922,
      "grad_norm": 20.151453018188477,
      "kl": 0.133544921875,
      "learning_rate": 4.75177304964539e-07,
      "loss": 0.0053,
      "reward": 1.766117811203003,
      "reward_std": 0.08813095837831497,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7661179304122925,
      "step": 1628
    },
    {
      "completion_length": 106.671875,
      "epoch": 1.0502901353965184,
      "grad_norm": 25.65150260925293,
      "kl": 0.12939453125,
      "learning_rate": 4.7485493230174076e-07,
      "loss": 0.0052,
      "reward": 1.480661690235138,
      "reward_std": 0.07892773300409317,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48066161572933197,
      "step": 1629
    },
    {
      "completion_length": 102.328125,
      "epoch": 1.0509348807221148,
      "grad_norm": 42.69692611694336,
      "kl": 0.13232421875,
      "learning_rate": 4.745325596389426e-07,
      "loss": 0.0053,
      "reward": 1.357925534248352,
      "reward_std": 0.07967722043395042,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.35792553424835205,
      "step": 1630
    },
    {
      "completion_length": 91.40625,
      "epoch": 1.051579626047711,
      "grad_norm": 24.92087173461914,
      "kl": 0.130859375,
      "learning_rate": 4.742101869761444e-07,
      "loss": 0.0052,
      "reward": 1.6058239936828613,
      "reward_std": 0.12468841299414635,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6058240830898285,
      "step": 1631
    },
    {
      "completion_length": 108.375,
      "epoch": 1.0522243713733075,
      "grad_norm": 18.37355613708496,
      "kl": 0.1162109375,
      "learning_rate": 4.7388781431334616e-07,
      "loss": 0.0047,
      "reward": 1.6409690976142883,
      "reward_std": 0.07419593259692192,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6409690976142883,
      "step": 1632
    },
    {
      "completion_length": 102.71875,
      "epoch": 1.052869116698904,
      "grad_norm": 13.083841323852539,
      "kl": 0.1337890625,
      "learning_rate": 4.73565441650548e-07,
      "loss": 0.0054,
      "reward": 1.5053048133850098,
      "reward_std": 0.099861279129982,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.520929828286171,
      "step": 1633
    },
    {
      "completion_length": 102.90625,
      "epoch": 1.0535138620245004,
      "grad_norm": 11.976901054382324,
      "kl": 0.141845703125,
      "learning_rate": 4.7324306898774977e-07,
      "loss": 0.0057,
      "reward": 1.6774804592132568,
      "reward_std": 0.11440124735236168,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6774804592132568,
      "step": 1634
    },
    {
      "completion_length": 114.65625,
      "epoch": 1.0541586073500968,
      "grad_norm": 13.442267417907715,
      "kl": 0.142578125,
      "learning_rate": 4.729206963249516e-07,
      "loss": 0.0057,
      "reward": 1.5921526551246643,
      "reward_std": 0.07790956273674965,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5921527147293091,
      "step": 1635
    },
    {
      "completion_length": 114.875,
      "epoch": 1.054803352675693,
      "grad_norm": 15.106107711791992,
      "kl": 0.13720703125,
      "learning_rate": 4.725983236621534e-07,
      "loss": 0.0055,
      "reward": 1.6467521786689758,
      "reward_std": 0.09242963790893555,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6467522084712982,
      "step": 1636
    },
    {
      "completion_length": 104.703125,
      "epoch": 1.0554480980012895,
      "grad_norm": 14.148865699768066,
      "kl": 0.114013671875,
      "learning_rate": 4.7227595099935527e-07,
      "loss": 0.0046,
      "reward": 1.7142847776412964,
      "reward_std": 0.08839049190282822,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7142847776412964,
      "step": 1637
    },
    {
      "completion_length": 93.859375,
      "epoch": 1.056092843326886,
      "grad_norm": 9.402307510375977,
      "kl": 0.15234375,
      "learning_rate": 4.71953578336557e-07,
      "loss": 0.0061,
      "reward": 1.6899645328521729,
      "reward_std": 0.09697774797677994,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6899645030498505,
      "step": 1638
    },
    {
      "completion_length": 110.109375,
      "epoch": 1.0567375886524824,
      "grad_norm": 25.695877075195312,
      "kl": 0.1630859375,
      "learning_rate": 4.716312056737589e-07,
      "loss": 0.0065,
      "reward": 1.5348122119903564,
      "reward_std": 0.09680547192692757,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5348122417926788,
      "step": 1639
    },
    {
      "completion_length": 116.484375,
      "epoch": 1.0573823339780786,
      "grad_norm": 26.15424346923828,
      "kl": 0.12939453125,
      "learning_rate": 4.7130883301096066e-07,
      "loss": 0.0052,
      "reward": 1.542705476284027,
      "reward_std": 0.09862326830625534,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5427054464817047,
      "step": 1640
    },
    {
      "completion_length": 111.40625,
      "epoch": 1.058027079303675,
      "grad_norm": 14.094022750854492,
      "kl": 0.12744140625,
      "learning_rate": 4.709864603481625e-07,
      "loss": 0.0051,
      "reward": 1.4890643954277039,
      "reward_std": 0.06961235404014587,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48906436562538147,
      "step": 1641
    },
    {
      "completion_length": 115.34375,
      "epoch": 1.0586718246292715,
      "grad_norm": 18.5950984954834,
      "kl": 0.122802734375,
      "learning_rate": 4.706640876853643e-07,
      "loss": 0.0049,
      "reward": 1.5355899930000305,
      "reward_std": 0.0737343579530716,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5355899631977081,
      "step": 1642
    },
    {
      "completion_length": 119.015625,
      "epoch": 1.059316569954868,
      "grad_norm": 10.859661102294922,
      "kl": 0.154296875,
      "learning_rate": 4.703417150225661e-07,
      "loss": 0.0062,
      "reward": 1.5704774856567383,
      "reward_std": 0.12773460149765015,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5704774707555771,
      "step": 1643
    },
    {
      "completion_length": 103.328125,
      "epoch": 1.0599613152804641,
      "grad_norm": 11.223685264587402,
      "kl": 0.130615234375,
      "learning_rate": 4.700193423597679e-07,
      "loss": 0.0052,
      "reward": 1.6081100702285767,
      "reward_std": 0.08056379854679108,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.608110100030899,
      "step": 1644
    },
    {
      "completion_length": 106.625,
      "epoch": 1.0606060606060606,
      "grad_norm": 49.92131805419922,
      "kl": 0.130126953125,
      "learning_rate": 4.696969696969697e-07,
      "loss": 0.0052,
      "reward": 1.4988184571266174,
      "reward_std": 0.09734079986810684,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49881842732429504,
      "step": 1645
    },
    {
      "completion_length": 115.0,
      "epoch": 1.061250805931657,
      "grad_norm": 30.047075271606445,
      "kl": 0.136474609375,
      "learning_rate": 4.693745970341715e-07,
      "loss": 0.0055,
      "reward": 1.6277170777320862,
      "reward_std": 0.04251527041196823,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6277171224355698,
      "step": 1646
    },
    {
      "completion_length": 110.390625,
      "epoch": 1.0618955512572534,
      "grad_norm": 8.765604972839355,
      "kl": 0.1376953125,
      "learning_rate": 4.690522243713733e-07,
      "loss": 0.0055,
      "reward": 1.5230787992477417,
      "reward_std": 0.09076399356126785,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5230787694454193,
      "step": 1647
    },
    {
      "completion_length": 117.78125,
      "epoch": 1.0625402965828499,
      "grad_norm": 18.16217613220215,
      "kl": 0.14111328125,
      "learning_rate": 4.687298517085751e-07,
      "loss": 0.0057,
      "reward": 1.4952101111412048,
      "reward_std": 0.1261536367237568,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49521011114120483,
      "step": 1648
    },
    {
      "completion_length": 102.578125,
      "epoch": 1.063185041908446,
      "grad_norm": 9.8458251953125,
      "kl": 0.13525390625,
      "learning_rate": 4.684074790457769e-07,
      "loss": 0.0054,
      "reward": 1.5231833457946777,
      "reward_std": 0.10076124221086502,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5388083457946777,
      "step": 1649
    },
    {
      "completion_length": 103.609375,
      "epoch": 1.0638297872340425,
      "grad_norm": 54.941654205322266,
      "kl": 0.12646484375,
      "learning_rate": 4.6808510638297873e-07,
      "loss": 0.0051,
      "reward": 1.5805718898773193,
      "reward_std": 0.11327047273516655,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.580571860074997,
      "step": 1650
    },
    {
      "completion_length": 121.703125,
      "epoch": 1.064474532559639,
      "grad_norm": 12.24212646484375,
      "kl": 0.14794921875,
      "learning_rate": 4.677627337201805e-07,
      "loss": 0.0059,
      "reward": 1.5812976360321045,
      "reward_std": 0.07798319682478905,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5812976062297821,
      "step": 1651
    },
    {
      "completion_length": 103.625,
      "epoch": 1.0651192778852354,
      "grad_norm": 7.415297985076904,
      "kl": 0.127685546875,
      "learning_rate": 4.6744036105738235e-07,
      "loss": 0.0051,
      "reward": 1.4523885846138,
      "reward_std": 0.12561476044356823,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45238858461380005,
      "step": 1652
    },
    {
      "completion_length": 115.21875,
      "epoch": 1.0657640232108316,
      "grad_norm": 12.325562477111816,
      "kl": 0.1337890625,
      "learning_rate": 4.6711798839458413e-07,
      "loss": 0.0054,
      "reward": 1.5610628724098206,
      "reward_std": 0.07157477736473083,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5610628575086594,
      "step": 1653
    },
    {
      "completion_length": 104.09375,
      "epoch": 1.066408768536428,
      "grad_norm": 20.869457244873047,
      "kl": 0.13818359375,
      "learning_rate": 4.6679561573178596e-07,
      "loss": 0.0055,
      "reward": 1.5744282603263855,
      "reward_std": 0.09144582599401474,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5744282603263855,
      "step": 1654
    },
    {
      "completion_length": 104.421875,
      "epoch": 1.0670535138620245,
      "grad_norm": 7.075346946716309,
      "kl": 0.14599609375,
      "learning_rate": 4.6647324306898774e-07,
      "loss": 0.0059,
      "reward": 1.6107639074325562,
      "reward_std": 0.13972818106412888,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6263889372348785,
      "step": 1655
    },
    {
      "completion_length": 100.640625,
      "epoch": 1.067698259187621,
      "grad_norm": 6.115519046783447,
      "kl": 0.122314453125,
      "learning_rate": 4.661508704061896e-07,
      "loss": 0.0049,
      "reward": 1.5545591115951538,
      "reward_std": 0.053600821644067764,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.554559051990509,
      "step": 1656
    },
    {
      "completion_length": 108.640625,
      "epoch": 1.0683430045132172,
      "grad_norm": 12.812996864318848,
      "kl": 0.134765625,
      "learning_rate": 4.6582849774339136e-07,
      "loss": 0.0054,
      "reward": 1.6389289498329163,
      "reward_std": 0.07384452596306801,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6389289498329163,
      "step": 1657
    },
    {
      "completion_length": 124.5625,
      "epoch": 1.0689877498388136,
      "grad_norm": 10.919026374816895,
      "kl": 0.09814453125,
      "learning_rate": 4.655061250805932e-07,
      "loss": 0.0039,
      "reward": 1.5925003290176392,
      "reward_std": 0.057900771498680115,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5925003588199615,
      "step": 1658
    },
    {
      "completion_length": 114.0,
      "epoch": 1.06963249516441,
      "grad_norm": 33.34097671508789,
      "kl": 0.1142578125,
      "learning_rate": 4.6518375241779497e-07,
      "loss": 0.0046,
      "reward": 1.700951337814331,
      "reward_std": 0.11340216547250748,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7009514570236206,
      "step": 1659
    },
    {
      "completion_length": 107.953125,
      "epoch": 1.0702772404900065,
      "grad_norm": 30.96587371826172,
      "kl": 0.120849609375,
      "learning_rate": 4.6486137975499675e-07,
      "loss": 0.0048,
      "reward": 1.5911786556243896,
      "reward_std": 0.21220052242279053,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6068036258220673,
      "step": 1660
    },
    {
      "completion_length": 97.828125,
      "epoch": 1.070921985815603,
      "grad_norm": 13.614809036254883,
      "kl": 0.1171875,
      "learning_rate": 4.645390070921986e-07,
      "loss": 0.0047,
      "reward": 1.5786557793617249,
      "reward_std": 0.10805037245154381,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5786558091640472,
      "step": 1661
    },
    {
      "completion_length": 105.53125,
      "epoch": 1.0715667311411992,
      "grad_norm": 13.102092742919922,
      "kl": 0.11474609375,
      "learning_rate": 4.6421663442940036e-07,
      "loss": 0.0046,
      "reward": 1.7086985111236572,
      "reward_std": 0.05099846050143242,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7086984813213348,
      "step": 1662
    },
    {
      "completion_length": 95.953125,
      "epoch": 1.0722114764667956,
      "grad_norm": 12.318425178527832,
      "kl": 0.1298828125,
      "learning_rate": 4.638942617666022e-07,
      "loss": 0.0052,
      "reward": 1.563442349433899,
      "reward_std": 0.04831070266664028,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5634424388408661,
      "step": 1663
    },
    {
      "completion_length": 113.578125,
      "epoch": 1.072856221792392,
      "grad_norm": 17.619489669799805,
      "kl": 0.101806640625,
      "learning_rate": 4.63571889103804e-07,
      "loss": 0.0041,
      "reward": 1.6703379154205322,
      "reward_std": 0.08459489978849888,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6703379154205322,
      "step": 1664
    },
    {
      "completion_length": 110.671875,
      "epoch": 1.0735009671179885,
      "grad_norm": 26.056339263916016,
      "kl": 0.13623046875,
      "learning_rate": 4.632495164410058e-07,
      "loss": 0.0054,
      "reward": 1.7076152563095093,
      "reward_std": 0.09282954782247543,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.707615315914154,
      "step": 1665
    },
    {
      "completion_length": 107.46875,
      "epoch": 1.0741457124435847,
      "grad_norm": 9.738430976867676,
      "kl": 0.10888671875,
      "learning_rate": 4.629271437782076e-07,
      "loss": 0.0044,
      "reward": 1.595727562904358,
      "reward_std": 0.12758340686559677,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5957275629043579,
      "step": 1666
    },
    {
      "completion_length": 102.265625,
      "epoch": 1.0747904577691811,
      "grad_norm": 17.98639488220215,
      "kl": 0.114501953125,
      "learning_rate": 4.626047711154094e-07,
      "loss": 0.0046,
      "reward": 1.5077466368675232,
      "reward_std": 0.04883422330021858,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5077466517686844,
      "step": 1667
    },
    {
      "completion_length": 109.546875,
      "epoch": 1.0754352030947776,
      "grad_norm": 9.506091117858887,
      "kl": 0.10986328125,
      "learning_rate": 4.622823984526112e-07,
      "loss": 0.0044,
      "reward": 1.5453261137008667,
      "reward_std": 0.07154025323688984,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5453260838985443,
      "step": 1668
    },
    {
      "completion_length": 102.984375,
      "epoch": 1.076079948420374,
      "grad_norm": 20.088478088378906,
      "kl": 0.15771484375,
      "learning_rate": 4.6196002578981304e-07,
      "loss": 0.0063,
      "reward": 1.5990652441978455,
      "reward_std": 0.08672228083014488,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5990653038024902,
      "step": 1669
    },
    {
      "completion_length": 108.390625,
      "epoch": 1.0767246937459702,
      "grad_norm": 9.232168197631836,
      "kl": 0.1298828125,
      "learning_rate": 4.616376531270148e-07,
      "loss": 0.0052,
      "reward": 1.639800250530243,
      "reward_std": 0.1497473306953907,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6398002207279205,
      "step": 1670
    },
    {
      "completion_length": 119.8125,
      "epoch": 1.0773694390715667,
      "grad_norm": 14.667919158935547,
      "kl": 0.140625,
      "learning_rate": 4.6131528046421665e-07,
      "loss": 0.0056,
      "reward": 1.4386180639266968,
      "reward_std": 0.10323400981724262,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43861812353134155,
      "step": 1671
    },
    {
      "completion_length": 110.765625,
      "epoch": 1.0780141843971631,
      "grad_norm": 26.46577262878418,
      "kl": 0.14306640625,
      "learning_rate": 4.6099290780141843e-07,
      "loss": 0.0057,
      "reward": 1.4531315565109253,
      "reward_std": 0.1466679349541664,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4687565565109253,
      "step": 1672
    },
    {
      "completion_length": 103.078125,
      "epoch": 1.0786589297227596,
      "grad_norm": 72.6494369506836,
      "kl": 0.115478515625,
      "learning_rate": 4.606705351386202e-07,
      "loss": 0.0046,
      "reward": 1.620385468006134,
      "reward_std": 0.09875315055251122,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6203854531049728,
      "step": 1673
    },
    {
      "completion_length": 109.53125,
      "epoch": 1.0793036750483558,
      "grad_norm": 105.22774505615234,
      "kl": 0.13525390625,
      "learning_rate": 4.6034816247582205e-07,
      "loss": 0.0054,
      "reward": 1.5872658491134644,
      "reward_std": 0.07589317858219147,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5872658789157867,
      "step": 1674
    },
    {
      "completion_length": 107.203125,
      "epoch": 1.0799484203739522,
      "grad_norm": 16.956668853759766,
      "kl": 0.113525390625,
      "learning_rate": 4.6002578981302383e-07,
      "loss": 0.0045,
      "reward": 1.683825135231018,
      "reward_std": 0.15254256501793861,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6994501352310181,
      "step": 1675
    },
    {
      "completion_length": 102.78125,
      "epoch": 1.0805931656995487,
      "grad_norm": 33.67800521850586,
      "kl": 0.11572265625,
      "learning_rate": 4.5970341715022566e-07,
      "loss": 0.0046,
      "reward": 1.730326533317566,
      "reward_std": 0.1375247836112976,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7303265333175659,
      "step": 1676
    },
    {
      "completion_length": 99.21875,
      "epoch": 1.081237911025145,
      "grad_norm": 12.1163969039917,
      "kl": 0.119140625,
      "learning_rate": 4.5938104448742744e-07,
      "loss": 0.0048,
      "reward": 1.5533208847045898,
      "reward_std": 0.09048768505454063,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5533208549022675,
      "step": 1677
    },
    {
      "completion_length": 99.03125,
      "epoch": 1.0818826563507415,
      "grad_norm": 21.379323959350586,
      "kl": 0.14306640625,
      "learning_rate": 4.590586718246293e-07,
      "loss": 0.0057,
      "reward": 1.547052264213562,
      "reward_std": 0.09882994741201401,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.547052264213562,
      "step": 1678
    },
    {
      "completion_length": 100.734375,
      "epoch": 1.0825274016763378,
      "grad_norm": 16.001218795776367,
      "kl": 0.21875,
      "learning_rate": 4.5873629916183106e-07,
      "loss": 0.0088,
      "reward": 1.509481430053711,
      "reward_std": 0.20628438889980316,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5251064896583557,
      "step": 1679
    },
    {
      "completion_length": 94.640625,
      "epoch": 1.0831721470019342,
      "grad_norm": 19.575855255126953,
      "kl": 0.13134765625,
      "learning_rate": 4.584139264990329e-07,
      "loss": 0.0052,
      "reward": 1.5105572938919067,
      "reward_std": 0.06101445481181145,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5105573087930679,
      "step": 1680
    },
    {
      "completion_length": 106.015625,
      "epoch": 1.0838168923275306,
      "grad_norm": 15.92323112487793,
      "kl": 0.1083984375,
      "learning_rate": 4.5809155383623467e-07,
      "loss": 0.0043,
      "reward": 1.755763292312622,
      "reward_std": 0.0871877558529377,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7557632923126221,
      "step": 1681
    },
    {
      "completion_length": 95.5625,
      "epoch": 1.084461637653127,
      "grad_norm": 13.914878845214844,
      "kl": 0.12890625,
      "learning_rate": 4.577691811734365e-07,
      "loss": 0.0052,
      "reward": 1.485410749912262,
      "reward_std": 0.12811970710754395,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4854106605052948,
      "step": 1682
    },
    {
      "completion_length": 111.90625,
      "epoch": 1.0851063829787233,
      "grad_norm": 10.895196914672852,
      "kl": 0.1318359375,
      "learning_rate": 4.574468085106383e-07,
      "loss": 0.0053,
      "reward": 1.5880500078201294,
      "reward_std": 0.10918740928173065,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5880500674247742,
      "step": 1683
    },
    {
      "completion_length": 97.40625,
      "epoch": 1.0857511283043197,
      "grad_norm": 21.795761108398438,
      "kl": 0.120849609375,
      "learning_rate": 4.571244358478401e-07,
      "loss": 0.0048,
      "reward": 1.4776260256767273,
      "reward_std": 0.06880192831158638,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47762608528137207,
      "step": 1684
    },
    {
      "completion_length": 102.421875,
      "epoch": 1.0863958736299162,
      "grad_norm": 14.754842758178711,
      "kl": 0.1103515625,
      "learning_rate": 4.568020631850419e-07,
      "loss": 0.0044,
      "reward": 1.710328757762909,
      "reward_std": 0.06619776226580143,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7103287577629089,
      "step": 1685
    },
    {
      "completion_length": 105.65625,
      "epoch": 1.0870406189555126,
      "grad_norm": 9.40377426147461,
      "kl": 0.4111328125,
      "learning_rate": 4.564796905222437e-07,
      "loss": 0.0164,
      "reward": 1.6777933239936829,
      "reward_std": 0.2508092671632767,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.7090433239936829,
      "step": 1686
    },
    {
      "completion_length": 104.53125,
      "epoch": 1.0876853642811088,
      "grad_norm": 175.24464416503906,
      "kl": 0.12646484375,
      "learning_rate": 4.561573178594455e-07,
      "loss": 0.0051,
      "reward": 1.484681248664856,
      "reward_std": 0.07167026400566101,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48468117415905,
      "step": 1687
    },
    {
      "completion_length": 97.96875,
      "epoch": 1.0883301096067053,
      "grad_norm": 12.59119987487793,
      "kl": 0.119873046875,
      "learning_rate": 4.558349451966473e-07,
      "loss": 0.0048,
      "reward": 1.7193995118141174,
      "reward_std": 0.058623988181352615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.719399482011795,
      "step": 1688
    },
    {
      "completion_length": 103.1875,
      "epoch": 1.0889748549323017,
      "grad_norm": 12.258233070373535,
      "kl": 0.15087890625,
      "learning_rate": 4.555125725338491e-07,
      "loss": 0.006,
      "reward": 1.6564828157424927,
      "reward_std": 0.08070876821875572,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6564827561378479,
      "step": 1689
    },
    {
      "completion_length": 114.921875,
      "epoch": 1.0896196002578982,
      "grad_norm": 21.755619049072266,
      "kl": 0.112060546875,
      "learning_rate": 4.551901998710509e-07,
      "loss": 0.0045,
      "reward": 1.6646082401275635,
      "reward_std": 0.09645628184080124,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6646082699298859,
      "step": 1690
    },
    {
      "completion_length": 104.421875,
      "epoch": 1.0902643455834946,
      "grad_norm": 10.000032424926758,
      "kl": 0.14501953125,
      "learning_rate": 4.5486782720825274e-07,
      "loss": 0.0058,
      "reward": 1.708804428577423,
      "reward_std": 0.17115085572004318,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7244293987751007,
      "step": 1691
    },
    {
      "completion_length": 107.71875,
      "epoch": 1.0909090909090908,
      "grad_norm": 26.18584632873535,
      "kl": 0.1240234375,
      "learning_rate": 4.545454545454545e-07,
      "loss": 0.005,
      "reward": 1.7140477299690247,
      "reward_std": 0.10914264246821404,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7140477001667023,
      "step": 1692
    },
    {
      "completion_length": 100.3125,
      "epoch": 1.0915538362346873,
      "grad_norm": 12.087810516357422,
      "kl": 0.119384765625,
      "learning_rate": 4.5422308188265635e-07,
      "loss": 0.0048,
      "reward": 1.6223152875900269,
      "reward_std": 0.07956447452306747,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6223153173923492,
      "step": 1693
    },
    {
      "completion_length": 105.265625,
      "epoch": 1.0921985815602837,
      "grad_norm": 11.739949226379395,
      "kl": 0.14208984375,
      "learning_rate": 4.5390070921985813e-07,
      "loss": 0.0057,
      "reward": 1.5097407698631287,
      "reward_std": 0.09296748787164688,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5097408145666122,
      "step": 1694
    },
    {
      "completion_length": 104.625,
      "epoch": 1.0928433268858802,
      "grad_norm": 11.357398986816406,
      "kl": 0.12060546875,
      "learning_rate": 4.5357833655705997e-07,
      "loss": 0.0048,
      "reward": 1.7072781324386597,
      "reward_std": 0.15004675462841988,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7229031324386597,
      "step": 1695
    },
    {
      "completion_length": 114.4375,
      "epoch": 1.0934880722114764,
      "grad_norm": 9.150431632995605,
      "kl": 0.110107421875,
      "learning_rate": 4.5325596389426175e-07,
      "loss": 0.0044,
      "reward": 1.6482697129249573,
      "reward_std": 0.06808177940547466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6482696235179901,
      "step": 1696
    },
    {
      "completion_length": 108.765625,
      "epoch": 1.0941328175370728,
      "grad_norm": 26.6113224029541,
      "kl": 0.143310546875,
      "learning_rate": 4.529335912314636e-07,
      "loss": 0.0057,
      "reward": 1.5538044571876526,
      "reward_std": 0.105574119836092,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5694293677806854,
      "step": 1697
    },
    {
      "completion_length": 112.109375,
      "epoch": 1.0947775628626693,
      "grad_norm": 15.188982963562012,
      "kl": 0.1513671875,
      "learning_rate": 4.5261121856866536e-07,
      "loss": 0.006,
      "reward": 1.4535160660743713,
      "reward_std": 0.14586328715085983,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4535161405801773,
      "step": 1698
    },
    {
      "completion_length": 101.59375,
      "epoch": 1.0954223081882657,
      "grad_norm": 9.596237182617188,
      "kl": 0.150390625,
      "learning_rate": 4.522888459058672e-07,
      "loss": 0.006,
      "reward": 1.618313193321228,
      "reward_std": 0.20893344283103943,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.6495631337165833,
      "step": 1699
    },
    {
      "completion_length": 99.671875,
      "epoch": 1.096067053513862,
      "grad_norm": 14.38936996459961,
      "kl": 0.13916015625,
      "learning_rate": 4.51966473243069e-07,
      "loss": 0.0056,
      "reward": 1.6061973571777344,
      "reward_std": 0.10794469714164734,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.606197327375412,
      "step": 1700
    },
    {
      "completion_length": 96.375,
      "epoch": 1.0967117988394584,
      "grad_norm": 81.89236450195312,
      "kl": 0.1357421875,
      "learning_rate": 4.5164410058027076e-07,
      "loss": 0.0054,
      "reward": 1.7045114636421204,
      "reward_std": 0.19201814383268356,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7201364934444427,
      "step": 1701
    },
    {
      "completion_length": 113.984375,
      "epoch": 1.0973565441650548,
      "grad_norm": 19.061376571655273,
      "kl": 0.10986328125,
      "learning_rate": 4.513217279174726e-07,
      "loss": 0.0044,
      "reward": 1.6339808106422424,
      "reward_std": 0.05904205143451691,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6339808702468872,
      "step": 1702
    },
    {
      "completion_length": 103.984375,
      "epoch": 1.0980012894906512,
      "grad_norm": 8.031238555908203,
      "kl": 0.110595703125,
      "learning_rate": 4.5099935525467437e-07,
      "loss": 0.0044,
      "reward": 1.6934545040130615,
      "reward_std": 0.11994657665491104,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6934545040130615,
      "step": 1703
    },
    {
      "completion_length": 107.28125,
      "epoch": 1.0986460348162477,
      "grad_norm": 73.1088638305664,
      "kl": 0.116943359375,
      "learning_rate": 4.506769825918762e-07,
      "loss": 0.0047,
      "reward": 1.6883281469345093,
      "reward_std": 0.10029840841889381,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6883280873298645,
      "step": 1704
    },
    {
      "completion_length": 106.125,
      "epoch": 1.099290780141844,
      "grad_norm": 18.33809471130371,
      "kl": 0.13134765625,
      "learning_rate": 4.50354609929078e-07,
      "loss": 0.0053,
      "reward": 1.5655602812767029,
      "reward_std": 0.15298619121313095,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5811853408813477,
      "step": 1705
    },
    {
      "completion_length": 111.8125,
      "epoch": 1.0999355254674403,
      "grad_norm": 13.750594139099121,
      "kl": 0.12158203125,
      "learning_rate": 4.500322372662798e-07,
      "loss": 0.0049,
      "reward": 1.5528867840766907,
      "reward_std": 0.14944328367710114,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5685118287801743,
      "step": 1706
    },
    {
      "completion_length": 107.0625,
      "epoch": 1.1005802707930368,
      "grad_norm": 15.120196342468262,
      "kl": 0.129150390625,
      "learning_rate": 4.497098646034816e-07,
      "loss": 0.0052,
      "reward": 1.7107568383216858,
      "reward_std": 0.06326938793063164,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7107568383216858,
      "step": 1707
    },
    {
      "completion_length": 97.53125,
      "epoch": 1.1012250161186332,
      "grad_norm": 37.98473358154297,
      "kl": 0.173828125,
      "learning_rate": 4.4938749194068343e-07,
      "loss": 0.007,
      "reward": 1.6928101778030396,
      "reward_std": 0.1691095195710659,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7084352076053619,
      "step": 1708
    },
    {
      "completion_length": 102.015625,
      "epoch": 1.1018697614442294,
      "grad_norm": 8.216543197631836,
      "kl": 0.13427734375,
      "learning_rate": 4.490651192778852e-07,
      "loss": 0.0054,
      "reward": 1.6665999293327332,
      "reward_std": 0.042136115953326225,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6665998697280884,
      "step": 1709
    },
    {
      "completion_length": 99.578125,
      "epoch": 1.1025145067698259,
      "grad_norm": 9.077375411987305,
      "kl": 0.13427734375,
      "learning_rate": 4.4874274661508705e-07,
      "loss": 0.0054,
      "reward": 1.7417653799057007,
      "reward_std": 0.12950639054179192,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7573904097080231,
      "step": 1710
    },
    {
      "completion_length": 98.71875,
      "epoch": 1.1031592520954223,
      "grad_norm": 15.816248893737793,
      "kl": 0.1611328125,
      "learning_rate": 4.484203739522888e-07,
      "loss": 0.0064,
      "reward": 1.2563967108726501,
      "reward_std": 0.24838687479496002,
      "rewards/format_reward": 0.953125,
      "rewards/iou_timestamp_reward": 0.30327168107032776,
      "step": 1711
    },
    {
      "completion_length": 94.796875,
      "epoch": 1.1038039974210188,
      "grad_norm": 13.886948585510254,
      "kl": 0.1337890625,
      "learning_rate": 4.4809800128949066e-07,
      "loss": 0.0053,
      "reward": 1.7339037656784058,
      "reward_std": 0.0918034091591835,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7339037656784058,
      "step": 1712
    },
    {
      "completion_length": 101.265625,
      "epoch": 1.104448742746615,
      "grad_norm": 30.30232048034668,
      "kl": 0.141357421875,
      "learning_rate": 4.4777562862669244e-07,
      "loss": 0.0056,
      "reward": 1.5611588954925537,
      "reward_std": 0.12069892883300781,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5767838507890701,
      "step": 1713
    },
    {
      "completion_length": 102.4375,
      "epoch": 1.1050934880722114,
      "grad_norm": 16.694547653198242,
      "kl": 0.130859375,
      "learning_rate": 4.474532559638942e-07,
      "loss": 0.0052,
      "reward": 1.698163628578186,
      "reward_std": 0.20370149239897728,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.7294135689735413,
      "step": 1714
    },
    {
      "completion_length": 96.671875,
      "epoch": 1.1057382333978079,
      "grad_norm": 12.910335540771484,
      "kl": 0.1259765625,
      "learning_rate": 4.4713088330109605e-07,
      "loss": 0.005,
      "reward": 1.4640353918075562,
      "reward_std": 0.03386647813022137,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4640353322029114,
      "step": 1715
    },
    {
      "completion_length": 99.234375,
      "epoch": 1.1063829787234043,
      "grad_norm": 20.440353393554688,
      "kl": 0.134765625,
      "learning_rate": 4.4680851063829783e-07,
      "loss": 0.0054,
      "reward": 1.6791711449623108,
      "reward_std": 0.07557164318859577,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.679171234369278,
      "step": 1716
    },
    {
      "completion_length": 98.0625,
      "epoch": 1.1070277240490007,
      "grad_norm": 12.584556579589844,
      "kl": 0.150390625,
      "learning_rate": 4.4648613797549967e-07,
      "loss": 0.006,
      "reward": 1.495797336101532,
      "reward_std": 0.15810192748904228,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5114222764968872,
      "step": 1717
    },
    {
      "completion_length": 88.4375,
      "epoch": 1.107672469374597,
      "grad_norm": 23.811677932739258,
      "kl": 0.1435546875,
      "learning_rate": 4.4616376531270145e-07,
      "loss": 0.0058,
      "reward": 1.4199917912483215,
      "reward_std": 0.1605324186384678,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.4512418508529663,
      "step": 1718
    },
    {
      "completion_length": 102.265625,
      "epoch": 1.1083172147001934,
      "grad_norm": 8.892284393310547,
      "kl": 0.128662109375,
      "learning_rate": 4.458413926499033e-07,
      "loss": 0.0051,
      "reward": 1.5546639561653137,
      "reward_std": 0.2123618647456169,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5859139859676361,
      "step": 1719
    },
    {
      "completion_length": 101.90625,
      "epoch": 1.1089619600257898,
      "grad_norm": 9.575830459594727,
      "kl": 0.1396484375,
      "learning_rate": 4.4551901998710506e-07,
      "loss": 0.0056,
      "reward": 1.5857290029525757,
      "reward_std": 0.07718907855451107,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5857289731502533,
      "step": 1720
    },
    {
      "completion_length": 106.078125,
      "epoch": 1.1096067053513863,
      "grad_norm": 273.09173583984375,
      "kl": 0.1279296875,
      "learning_rate": 4.451966473243069e-07,
      "loss": 0.0051,
      "reward": 1.488102912902832,
      "reward_std": 0.17496028542518616,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5037279427051544,
      "step": 1721
    },
    {
      "completion_length": 111.328125,
      "epoch": 1.1102514506769825,
      "grad_norm": 15.99120044708252,
      "kl": 0.12451171875,
      "learning_rate": 4.448742746615087e-07,
      "loss": 0.005,
      "reward": 1.6247381567955017,
      "reward_std": 0.1117971558123827,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6247380971908569,
      "step": 1722
    },
    {
      "completion_length": 107.953125,
      "epoch": 1.110896196002579,
      "grad_norm": 9.210831642150879,
      "kl": 0.107666015625,
      "learning_rate": 4.445519019987105e-07,
      "loss": 0.0043,
      "reward": 1.5621591806411743,
      "reward_std": 0.18757861852645874,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5934091806411743,
      "step": 1723
    },
    {
      "completion_length": 106.078125,
      "epoch": 1.1115409413281754,
      "grad_norm": 39.334266662597656,
      "kl": 0.1162109375,
      "learning_rate": 4.442295293359123e-07,
      "loss": 0.0046,
      "reward": 1.404995858669281,
      "reward_std": 0.05613952688872814,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.40499579906463623,
      "step": 1724
    },
    {
      "completion_length": 114.71875,
      "epoch": 1.1121856866537718,
      "grad_norm": 15.110870361328125,
      "kl": 0.114501953125,
      "learning_rate": 4.439071566731141e-07,
      "loss": 0.0046,
      "reward": 1.5173662900924683,
      "reward_std": 0.05804556608200073,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5173663049936295,
      "step": 1725
    },
    {
      "completion_length": 102.265625,
      "epoch": 1.112830431979368,
      "grad_norm": 16.11209487915039,
      "kl": 0.140625,
      "learning_rate": 4.435847840103159e-07,
      "loss": 0.0056,
      "reward": 1.5029131174087524,
      "reward_std": 0.13497604429721832,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5029131919145584,
      "step": 1726
    },
    {
      "completion_length": 110.375,
      "epoch": 1.1134751773049645,
      "grad_norm": 19.925506591796875,
      "kl": 0.121826171875,
      "learning_rate": 4.432624113475177e-07,
      "loss": 0.0049,
      "reward": 1.668049395084381,
      "reward_std": 0.06225564517080784,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6680494248867035,
      "step": 1727
    },
    {
      "completion_length": 99.625,
      "epoch": 1.114119922630561,
      "grad_norm": 12.299700736999512,
      "kl": 0.1357421875,
      "learning_rate": 4.429400386847195e-07,
      "loss": 0.0054,
      "reward": 1.632355809211731,
      "reward_std": 0.1639389768242836,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.6636057496070862,
      "step": 1728
    },
    {
      "completion_length": 109.71875,
      "epoch": 1.1147646679561574,
      "grad_norm": 24.155738830566406,
      "kl": 0.1142578125,
      "learning_rate": 4.426176660219213e-07,
      "loss": 0.0046,
      "reward": 1.5948566198349,
      "reward_std": 0.08682410418987274,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5948565602302551,
      "step": 1729
    },
    {
      "completion_length": 120.21875,
      "epoch": 1.1154094132817538,
      "grad_norm": 6.264504909515381,
      "kl": 0.1767578125,
      "learning_rate": 4.4229529335912313e-07,
      "loss": 0.0071,
      "reward": 1.5787187814712524,
      "reward_std": 0.08336924016475677,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.578718900680542,
      "step": 1730
    },
    {
      "completion_length": 115.609375,
      "epoch": 1.11605415860735,
      "grad_norm": 17.270288467407227,
      "kl": 0.1083984375,
      "learning_rate": 4.419729206963249e-07,
      "loss": 0.0043,
      "reward": 1.6184866428375244,
      "reward_std": 0.17556436359882355,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.6497366428375244,
      "step": 1731
    },
    {
      "completion_length": 116.796875,
      "epoch": 1.1166989039329465,
      "grad_norm": 15.003190994262695,
      "kl": 0.12255859375,
      "learning_rate": 4.4165054803352675e-07,
      "loss": 0.0049,
      "reward": 1.5859253406524658,
      "reward_std": 0.11098824068903923,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6015503257513046,
      "step": 1732
    },
    {
      "completion_length": 119.15625,
      "epoch": 1.117343649258543,
      "grad_norm": 11.927645683288574,
      "kl": 0.11328125,
      "learning_rate": 4.413281753707285e-07,
      "loss": 0.0045,
      "reward": 1.707281231880188,
      "reward_std": 0.09659011662006378,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.707281231880188,
      "step": 1733
    },
    {
      "completion_length": 108.671875,
      "epoch": 1.1179883945841393,
      "grad_norm": 11.922825813293457,
      "kl": 0.118408203125,
      "learning_rate": 4.4100580270793036e-07,
      "loss": 0.0047,
      "reward": 1.5895445942878723,
      "reward_std": 0.053144171833992004,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5895446836948395,
      "step": 1734
    },
    {
      "completion_length": 106.84375,
      "epoch": 1.1186331399097356,
      "grad_norm": 14.034700393676758,
      "kl": 0.11572265625,
      "learning_rate": 4.4068343004513214e-07,
      "loss": 0.0046,
      "reward": 1.6887595057487488,
      "reward_std": 0.10110663995146751,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6887595057487488,
      "step": 1735
    },
    {
      "completion_length": 119.546875,
      "epoch": 1.119277885235332,
      "grad_norm": 21.64225959777832,
      "kl": 0.12548828125,
      "learning_rate": 4.4036105738233397e-07,
      "loss": 0.005,
      "reward": 1.6815902590751648,
      "reward_std": 0.10987572371959686,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6815903186798096,
      "step": 1736
    },
    {
      "completion_length": 108.0625,
      "epoch": 1.1199226305609284,
      "grad_norm": 9.544087409973145,
      "kl": 0.1376953125,
      "learning_rate": 4.4003868471953575e-07,
      "loss": 0.0055,
      "reward": 1.6549758315086365,
      "reward_std": 0.08593714982271194,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6549758017063141,
      "step": 1737
    },
    {
      "completion_length": 121.21875,
      "epoch": 1.1205673758865249,
      "grad_norm": 39.510986328125,
      "kl": 0.10791015625,
      "learning_rate": 4.397163120567376e-07,
      "loss": 0.0043,
      "reward": 1.6081666350364685,
      "reward_std": 0.09677704423666,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6081666350364685,
      "step": 1738
    },
    {
      "completion_length": 116.953125,
      "epoch": 1.121212121212121,
      "grad_norm": 9.956169128417969,
      "kl": 0.134765625,
      "learning_rate": 4.3939393939393937e-07,
      "loss": 0.0054,
      "reward": 1.4132041931152344,
      "reward_std": 0.09558882936835289,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.428829163312912,
      "step": 1739
    },
    {
      "completion_length": 117.0,
      "epoch": 1.1218568665377175,
      "grad_norm": 19.143388748168945,
      "kl": 0.102294921875,
      "learning_rate": 4.3907156673114115e-07,
      "loss": 0.0041,
      "reward": 1.5230024456977844,
      "reward_std": 0.07811861485242844,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5230023860931396,
      "step": 1740
    },
    {
      "completion_length": 110.5625,
      "epoch": 1.122501611863314,
      "grad_norm": 17.286823272705078,
      "kl": 0.115234375,
      "learning_rate": 4.38749194068343e-07,
      "loss": 0.0046,
      "reward": 1.6897668838500977,
      "reward_std": 0.09538877755403519,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6897669434547424,
      "step": 1741
    },
    {
      "completion_length": 123.3125,
      "epoch": 1.1231463571889104,
      "grad_norm": 15.611011505126953,
      "kl": 0.1240234375,
      "learning_rate": 4.3842682140554476e-07,
      "loss": 0.005,
      "reward": 1.457338809967041,
      "reward_std": 0.1063726358115673,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4573388397693634,
      "step": 1742
    },
    {
      "completion_length": 116.84375,
      "epoch": 1.1237911025145069,
      "grad_norm": 14.022031784057617,
      "kl": 0.115966796875,
      "learning_rate": 4.381044487427466e-07,
      "loss": 0.0046,
      "reward": 1.3342976570129395,
      "reward_std": 0.08087656646966934,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.33429761230945587,
      "step": 1743
    },
    {
      "completion_length": 113.390625,
      "epoch": 1.124435847840103,
      "grad_norm": 9.711297988891602,
      "kl": 0.102294921875,
      "learning_rate": 4.377820760799484e-07,
      "loss": 0.0041,
      "reward": 1.548888087272644,
      "reward_std": 0.09965697675943375,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5488881766796112,
      "step": 1744
    },
    {
      "completion_length": 116.15625,
      "epoch": 1.1250805931656995,
      "grad_norm": 24.13825225830078,
      "kl": 0.130859375,
      "learning_rate": 4.374597034171502e-07,
      "loss": 0.0052,
      "reward": 1.7484040260314941,
      "reward_std": 0.12463311851024628,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7484039068222046,
      "step": 1745
    },
    {
      "completion_length": 118.625,
      "epoch": 1.125725338491296,
      "grad_norm": 15.037027359008789,
      "kl": 0.10888671875,
      "learning_rate": 4.37137330754352e-07,
      "loss": 0.0044,
      "reward": 1.6446164846420288,
      "reward_std": 0.12209159508347511,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6602414548397064,
      "step": 1746
    },
    {
      "completion_length": 115.640625,
      "epoch": 1.1263700838168924,
      "grad_norm": 19.545330047607422,
      "kl": 0.119140625,
      "learning_rate": 4.368149580915538e-07,
      "loss": 0.0048,
      "reward": 1.4536727666854858,
      "reward_std": 0.17526058107614517,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4692978411912918,
      "step": 1747
    },
    {
      "completion_length": 106.40625,
      "epoch": 1.1270148291424886,
      "grad_norm": 18.563114166259766,
      "kl": 0.119140625,
      "learning_rate": 4.364925854287556e-07,
      "loss": 0.0048,
      "reward": 1.6942253708839417,
      "reward_std": 0.07673277705907822,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6942253112792969,
      "step": 1748
    },
    {
      "completion_length": 121.90625,
      "epoch": 1.127659574468085,
      "grad_norm": 12.64437198638916,
      "kl": 0.107421875,
      "learning_rate": 4.3617021276595744e-07,
      "loss": 0.0043,
      "reward": 1.4421557188034058,
      "reward_std": 0.11760647222399712,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4421556890010834,
      "step": 1749
    },
    {
      "completion_length": 111.46875,
      "epoch": 1.1283043197936815,
      "grad_norm": 25.638565063476562,
      "kl": 0.1650390625,
      "learning_rate": 4.358478401031592e-07,
      "loss": 0.0066,
      "reward": 1.6255099177360535,
      "reward_std": 0.1136264055967331,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6255099177360535,
      "step": 1750
    },
    {
      "completion_length": 105.421875,
      "epoch": 1.128949065119278,
      "grad_norm": 12.374931335449219,
      "kl": 0.127685546875,
      "learning_rate": 4.3552546744036105e-07,
      "loss": 0.0051,
      "reward": 1.6772376894950867,
      "reward_std": 0.1765252035111189,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.7084877490997314,
      "step": 1751
    },
    {
      "completion_length": 113.078125,
      "epoch": 1.1295938104448742,
      "grad_norm": 14.301130294799805,
      "kl": 0.12744140625,
      "learning_rate": 4.3520309477756283e-07,
      "loss": 0.0051,
      "reward": 1.494019091129303,
      "reward_std": 0.11037448793649673,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5096440762281418,
      "step": 1752
    },
    {
      "completion_length": 116.546875,
      "epoch": 1.1302385557704706,
      "grad_norm": 13.462944030761719,
      "kl": 0.205078125,
      "learning_rate": 4.3488072211476467e-07,
      "loss": 0.0082,
      "reward": 1.5543888211250305,
      "reward_std": 0.08765115961432457,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5543888360261917,
      "step": 1753
    },
    {
      "completion_length": 103.859375,
      "epoch": 1.130883301096067,
      "grad_norm": 58.70383071899414,
      "kl": 0.13525390625,
      "learning_rate": 4.3455834945196645e-07,
      "loss": 0.0054,
      "reward": 1.5313851833343506,
      "reward_std": 0.14955976232886314,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5313852429389954,
      "step": 1754
    },
    {
      "completion_length": 103.203125,
      "epoch": 1.1315280464216635,
      "grad_norm": 10.312337875366211,
      "kl": 0.11767578125,
      "learning_rate": 4.342359767891682e-07,
      "loss": 0.0047,
      "reward": 1.536129117012024,
      "reward_std": 0.11497850716114044,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5517541468143463,
      "step": 1755
    },
    {
      "completion_length": 99.140625,
      "epoch": 1.13217279174726,
      "grad_norm": 10.849982261657715,
      "kl": 0.11181640625,
      "learning_rate": 4.3391360412637006e-07,
      "loss": 0.0045,
      "reward": 1.7079610228538513,
      "reward_std": 0.10665994510054588,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7079610228538513,
      "step": 1756
    },
    {
      "completion_length": 106.9375,
      "epoch": 1.1328175370728562,
      "grad_norm": 10.794480323791504,
      "kl": 0.108154296875,
      "learning_rate": 4.3359123146357184e-07,
      "loss": 0.0043,
      "reward": 1.5266578197479248,
      "reward_std": 0.10716821625828743,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5422827899456024,
      "step": 1757
    },
    {
      "completion_length": 98.09375,
      "epoch": 1.1334622823984526,
      "grad_norm": 31.109466552734375,
      "kl": 0.1396484375,
      "learning_rate": 4.3326885880077367e-07,
      "loss": 0.0056,
      "reward": 1.6419140696525574,
      "reward_std": 0.11855699494481087,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6419140696525574,
      "step": 1758
    },
    {
      "completion_length": 109.0,
      "epoch": 1.134107027724049,
      "grad_norm": 11.9324312210083,
      "kl": 0.12060546875,
      "learning_rate": 4.3294648613797545e-07,
      "loss": 0.0048,
      "reward": 1.6172628998756409,
      "reward_std": 0.09370382130146027,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6172628998756409,
      "step": 1759
    },
    {
      "completion_length": 107.703125,
      "epoch": 1.1347517730496455,
      "grad_norm": 8.980454444885254,
      "kl": 0.107666015625,
      "learning_rate": 4.326241134751773e-07,
      "loss": 0.0043,
      "reward": 1.5437541007995605,
      "reward_std": 0.09230579063296318,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5437541604042053,
      "step": 1760
    },
    {
      "completion_length": 103.984375,
      "epoch": 1.1353965183752417,
      "grad_norm": 16.976593017578125,
      "kl": 0.119384765625,
      "learning_rate": 4.3230174081237907e-07,
      "loss": 0.0048,
      "reward": 1.6876962184906006,
      "reward_std": 0.1564544513821602,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6876962184906006,
      "step": 1761
    },
    {
      "completion_length": 96.140625,
      "epoch": 1.1360412637008381,
      "grad_norm": 16.373489379882812,
      "kl": 0.12060546875,
      "learning_rate": 4.319793681495809e-07,
      "loss": 0.0048,
      "reward": 1.7533943057060242,
      "reward_std": 0.11407231912016869,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7533943355083466,
      "step": 1762
    },
    {
      "completion_length": 104.96875,
      "epoch": 1.1366860090264346,
      "grad_norm": 12.83205509185791,
      "kl": 0.12841796875,
      "learning_rate": 4.316569954867827e-07,
      "loss": 0.0051,
      "reward": 1.7652117609977722,
      "reward_std": 0.11743732914328575,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7652117908000946,
      "step": 1763
    },
    {
      "completion_length": 106.28125,
      "epoch": 1.137330754352031,
      "grad_norm": 15.829553604125977,
      "kl": 0.130126953125,
      "learning_rate": 4.313346228239845e-07,
      "loss": 0.0052,
      "reward": 1.6745178699493408,
      "reward_std": 0.11732401326298714,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6745178401470184,
      "step": 1764
    },
    {
      "completion_length": 104.984375,
      "epoch": 1.1379754996776272,
      "grad_norm": 31.47109031677246,
      "kl": 0.13427734375,
      "learning_rate": 4.310122501611863e-07,
      "loss": 0.0054,
      "reward": 1.5751590132713318,
      "reward_std": 0.17913639172911644,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5907839834690094,
      "step": 1765
    },
    {
      "completion_length": 104.890625,
      "epoch": 1.1386202450032237,
      "grad_norm": 14.089750289916992,
      "kl": 0.124755859375,
      "learning_rate": 4.3068987749838813e-07,
      "loss": 0.005,
      "reward": 1.617997169494629,
      "reward_std": 0.06984603218734264,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6179972290992737,
      "step": 1766
    },
    {
      "completion_length": 98.046875,
      "epoch": 1.1392649903288201,
      "grad_norm": 52.08578872680664,
      "kl": 0.14697265625,
      "learning_rate": 4.303675048355899e-07,
      "loss": 0.0059,
      "reward": 1.4860327243804932,
      "reward_std": 0.05698714032769203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48603275418281555,
      "step": 1767
    },
    {
      "completion_length": 97.234375,
      "epoch": 1.1399097356544166,
      "grad_norm": 13.764451026916504,
      "kl": 0.1318359375,
      "learning_rate": 4.300451321727917e-07,
      "loss": 0.0053,
      "reward": 1.4116371273994446,
      "reward_std": 0.15613991767168045,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4272621124982834,
      "step": 1768
    },
    {
      "completion_length": 93.375,
      "epoch": 1.140554480980013,
      "grad_norm": 21.74166488647461,
      "kl": 0.114990234375,
      "learning_rate": 4.297227595099935e-07,
      "loss": 0.0046,
      "reward": 1.5942260026931763,
      "reward_std": 0.13864929229021072,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5942259430885315,
      "step": 1769
    },
    {
      "completion_length": 97.78125,
      "epoch": 1.1411992263056092,
      "grad_norm": 17.24546241760254,
      "kl": 0.11474609375,
      "learning_rate": 4.294003868471953e-07,
      "loss": 0.0046,
      "reward": 1.6961501836776733,
      "reward_std": 0.10894617810845375,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7117751836776733,
      "step": 1770
    },
    {
      "completion_length": 97.21875,
      "epoch": 1.1418439716312057,
      "grad_norm": 23.654821395874023,
      "kl": 0.143310546875,
      "learning_rate": 4.2907801418439714e-07,
      "loss": 0.0057,
      "reward": 1.6986296772956848,
      "reward_std": 0.10340171307325363,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6986297369003296,
      "step": 1771
    },
    {
      "completion_length": 97.625,
      "epoch": 1.142488716956802,
      "grad_norm": 16.723711013793945,
      "kl": 0.1376953125,
      "learning_rate": 4.287556415215989e-07,
      "loss": 0.0055,
      "reward": 1.6945127248764038,
      "reward_std": 0.10590310022234917,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.694512665271759,
      "step": 1772
    },
    {
      "completion_length": 102.0,
      "epoch": 1.1431334622823985,
      "grad_norm": 11.695935249328613,
      "kl": 0.1123046875,
      "learning_rate": 4.2843326885880075e-07,
      "loss": 0.0045,
      "reward": 1.6573434472084045,
      "reward_std": 0.12627606093883514,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6573435068130493,
      "step": 1773
    },
    {
      "completion_length": 101.734375,
      "epoch": 1.1437782076079948,
      "grad_norm": 11.529833793640137,
      "kl": 0.150390625,
      "learning_rate": 4.2811089619600253e-07,
      "loss": 0.006,
      "reward": 1.5360539555549622,
      "reward_std": 0.1177862137556076,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5360539853572845,
      "step": 1774
    },
    {
      "completion_length": 103.546875,
      "epoch": 1.1444229529335912,
      "grad_norm": 15.755107879638672,
      "kl": 0.19140625,
      "learning_rate": 4.2778852353320437e-07,
      "loss": 0.0076,
      "reward": 1.6712934970855713,
      "reward_std": 0.21360860764980316,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6869185268878937,
      "step": 1775
    },
    {
      "completion_length": 115.65625,
      "epoch": 1.1450676982591876,
      "grad_norm": 12.252145767211914,
      "kl": 0.12109375,
      "learning_rate": 4.2746615087040615e-07,
      "loss": 0.0048,
      "reward": 1.4710392951965332,
      "reward_std": 0.0702495351433754,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4710392951965332,
      "step": 1776
    },
    {
      "completion_length": 108.609375,
      "epoch": 1.145712443584784,
      "grad_norm": 11.388358116149902,
      "kl": 0.117431640625,
      "learning_rate": 4.27143778207608e-07,
      "loss": 0.0047,
      "reward": 1.7197344303131104,
      "reward_std": 0.09033240005373955,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7197344303131104,
      "step": 1777
    },
    {
      "completion_length": 115.328125,
      "epoch": 1.1463571889103803,
      "grad_norm": 10.445749282836914,
      "kl": 0.1025390625,
      "learning_rate": 4.2682140554480976e-07,
      "loss": 0.0041,
      "reward": 1.6068493723869324,
      "reward_std": 0.11387959867715836,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6068494319915771,
      "step": 1778
    },
    {
      "completion_length": 107.09375,
      "epoch": 1.1470019342359767,
      "grad_norm": 8.58862018585205,
      "kl": 0.125,
      "learning_rate": 4.264990328820116e-07,
      "loss": 0.005,
      "reward": 1.57830011844635,
      "reward_std": 0.145315021276474,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5783001780509949,
      "step": 1779
    },
    {
      "completion_length": 110.890625,
      "epoch": 1.1476466795615732,
      "grad_norm": 12.297728538513184,
      "kl": 0.11572265625,
      "learning_rate": 4.2617666021921337e-07,
      "loss": 0.0046,
      "reward": 1.5999037027359009,
      "reward_std": 0.10857437551021576,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5999037325382233,
      "step": 1780
    },
    {
      "completion_length": 102.65625,
      "epoch": 1.1482914248871696,
      "grad_norm": 32.0905647277832,
      "kl": 0.12451171875,
      "learning_rate": 4.2585428755641515e-07,
      "loss": 0.005,
      "reward": 1.7533397674560547,
      "reward_std": 0.06611281260848045,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7533398270606995,
      "step": 1781
    },
    {
      "completion_length": 98.78125,
      "epoch": 1.148936170212766,
      "grad_norm": 59.476600646972656,
      "kl": 0.14013671875,
      "learning_rate": 4.25531914893617e-07,
      "loss": 0.0056,
      "reward": 1.6604995727539062,
      "reward_std": 0.06654183939099312,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6604995727539062,
      "step": 1782
    },
    {
      "completion_length": 108.09375,
      "epoch": 1.1495809155383623,
      "grad_norm": 8.271899223327637,
      "kl": 0.120849609375,
      "learning_rate": 4.2520954223081877e-07,
      "loss": 0.0048,
      "reward": 1.637656807899475,
      "reward_std": 0.0554551612585783,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6376567780971527,
      "step": 1783
    },
    {
      "completion_length": 111.640625,
      "epoch": 1.1502256608639587,
      "grad_norm": 16.318239212036133,
      "kl": 0.1220703125,
      "learning_rate": 4.248871695680206e-07,
      "loss": 0.0049,
      "reward": 1.692155122756958,
      "reward_std": 0.16502898186445236,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.692155122756958,
      "step": 1784
    },
    {
      "completion_length": 106.0,
      "epoch": 1.1508704061895552,
      "grad_norm": 18.496143341064453,
      "kl": 0.110595703125,
      "learning_rate": 4.245647969052224e-07,
      "loss": 0.0044,
      "reward": 1.6564997434616089,
      "reward_std": 0.08594655059278011,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6564997732639313,
      "step": 1785
    },
    {
      "completion_length": 114.234375,
      "epoch": 1.1515151515151516,
      "grad_norm": 19.95158576965332,
      "kl": 0.1357421875,
      "learning_rate": 4.242424242424242e-07,
      "loss": 0.0054,
      "reward": 1.605454921722412,
      "reward_std": 0.08248751610517502,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6054548621177673,
      "step": 1786
    },
    {
      "completion_length": 103.53125,
      "epoch": 1.1521598968407478,
      "grad_norm": 29.761093139648438,
      "kl": 0.141357421875,
      "learning_rate": 4.23920051579626e-07,
      "loss": 0.0057,
      "reward": 1.6491164565086365,
      "reward_std": 0.15113864839076996,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6491163969039917,
      "step": 1787
    },
    {
      "completion_length": 95.421875,
      "epoch": 1.1528046421663443,
      "grad_norm": 27.836938858032227,
      "kl": 0.149658203125,
      "learning_rate": 4.2359767891682783e-07,
      "loss": 0.006,
      "reward": 1.5169153213500977,
      "reward_std": 0.12604781612753868,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5325403362512589,
      "step": 1788
    },
    {
      "completion_length": 110.484375,
      "epoch": 1.1534493874919407,
      "grad_norm": 20.547147750854492,
      "kl": 0.13330078125,
      "learning_rate": 4.232753062540296e-07,
      "loss": 0.0053,
      "reward": 1.5799216628074646,
      "reward_std": 0.07361920922994614,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.579921692609787,
      "step": 1789
    },
    {
      "completion_length": 106.234375,
      "epoch": 1.1540941328175371,
      "grad_norm": 34.002098083496094,
      "kl": 0.130859375,
      "learning_rate": 4.2295293359123144e-07,
      "loss": 0.0052,
      "reward": 1.6779032349586487,
      "reward_std": 0.11163284257054329,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6779031455516815,
      "step": 1790
    },
    {
      "completion_length": 109.5,
      "epoch": 1.1547388781431334,
      "grad_norm": 17.270278930664062,
      "kl": 0.123291015625,
      "learning_rate": 4.226305609284332e-07,
      "loss": 0.0049,
      "reward": 1.6175901889801025,
      "reward_std": 0.11164752580225468,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6175901889801025,
      "step": 1791
    },
    {
      "completion_length": 102.640625,
      "epoch": 1.1553836234687298,
      "grad_norm": 16.843664169311523,
      "kl": 0.1591796875,
      "learning_rate": 4.2230818826563506e-07,
      "loss": 0.0064,
      "reward": 1.7334028482437134,
      "reward_std": 0.07044504210352898,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.733402818441391,
      "step": 1792
    },
    {
      "completion_length": 113.375,
      "epoch": 1.1560283687943262,
      "grad_norm": 11.798425674438477,
      "kl": 0.130859375,
      "learning_rate": 4.2198581560283684e-07,
      "loss": 0.0052,
      "reward": 1.7188551425933838,
      "reward_std": 0.0985119491815567,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7188552021980286,
      "step": 1793
    },
    {
      "completion_length": 106.25,
      "epoch": 1.1566731141199227,
      "grad_norm": 15.71810531616211,
      "kl": 0.146484375,
      "learning_rate": 4.216634429400386e-07,
      "loss": 0.0059,
      "reward": 1.4141835570335388,
      "reward_std": 0.1422448419034481,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.41418349742889404,
      "step": 1794
    },
    {
      "completion_length": 104.65625,
      "epoch": 1.1573178594455191,
      "grad_norm": 28.754180908203125,
      "kl": 0.1455078125,
      "learning_rate": 4.2134107027724045e-07,
      "loss": 0.0058,
      "reward": 1.5761706233024597,
      "reward_std": 0.08441785722970963,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5761705487966537,
      "step": 1795
    },
    {
      "completion_length": 116.609375,
      "epoch": 1.1579626047711153,
      "grad_norm": 8.733136177062988,
      "kl": 0.12841796875,
      "learning_rate": 4.2101869761444223e-07,
      "loss": 0.0051,
      "reward": 1.5780669450759888,
      "reward_std": 0.08834883198142052,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5780669152736664,
      "step": 1796
    },
    {
      "completion_length": 124.0,
      "epoch": 1.1586073500967118,
      "grad_norm": 15.45083236694336,
      "kl": 0.13037109375,
      "learning_rate": 4.2069632495164407e-07,
      "loss": 0.0052,
      "reward": 1.6721486449241638,
      "reward_std": 0.23771092295646667,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.7033986151218414,
      "step": 1797
    },
    {
      "completion_length": 118.15625,
      "epoch": 1.1592520954223082,
      "grad_norm": 10.324441909790039,
      "kl": 0.1259765625,
      "learning_rate": 4.2037395228884585e-07,
      "loss": 0.005,
      "reward": 1.5779080986976624,
      "reward_std": 0.18352527171373367,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5935330986976624,
      "step": 1798
    },
    {
      "completion_length": 116.0625,
      "epoch": 1.1598968407479047,
      "grad_norm": 36.42701721191406,
      "kl": 0.112060546875,
      "learning_rate": 4.200515796260477e-07,
      "loss": 0.0045,
      "reward": 1.5790598392486572,
      "reward_std": 0.056727977469563484,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5790598094463348,
      "step": 1799
    },
    {
      "completion_length": 121.5625,
      "epoch": 1.1605415860735009,
      "grad_norm": 35.930091857910156,
      "kl": 0.111572265625,
      "learning_rate": 4.1972920696324946e-07,
      "loss": 0.0045,
      "reward": 1.608137845993042,
      "reward_std": 0.14337752014398575,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6081377863883972,
      "step": 1800
    },
    {
      "completion_length": 118.8125,
      "epoch": 1.1611863313990973,
      "grad_norm": 6.747325420379639,
      "kl": 0.134765625,
      "learning_rate": 4.194068343004513e-07,
      "loss": 0.0054,
      "reward": 1.5423181653022766,
      "reward_std": 0.08650622889399529,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.542318195104599,
      "step": 1801
    },
    {
      "completion_length": 110.59375,
      "epoch": 1.1618310767246938,
      "grad_norm": 11.091395378112793,
      "kl": 0.16552734375,
      "learning_rate": 4.190844616376531e-07,
      "loss": 0.0066,
      "reward": 1.6449681520462036,
      "reward_std": 0.10411114990711212,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.644968181848526,
      "step": 1802
    },
    {
      "completion_length": 111.71875,
      "epoch": 1.1624758220502902,
      "grad_norm": 14.368660926818848,
      "kl": 0.13525390625,
      "learning_rate": 4.187620889748549e-07,
      "loss": 0.0054,
      "reward": 1.7189124822616577,
      "reward_std": 0.09125639125704765,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7189124822616577,
      "step": 1803
    },
    {
      "completion_length": 124.25,
      "epoch": 1.1631205673758864,
      "grad_norm": 9.280939102172852,
      "kl": 0.12548828125,
      "learning_rate": 4.184397163120567e-07,
      "loss": 0.005,
      "reward": 1.5595988035202026,
      "reward_std": 0.09420382976531982,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5595987737178802,
      "step": 1804
    },
    {
      "completion_length": 103.65625,
      "epoch": 1.1637653127014829,
      "grad_norm": 27.450349807739258,
      "kl": 0.12939453125,
      "learning_rate": 4.181173436492585e-07,
      "loss": 0.0052,
      "reward": 1.6111841797828674,
      "reward_std": 0.1274169199168682,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.626809149980545,
      "step": 1805
    },
    {
      "completion_length": 107.65625,
      "epoch": 1.1644100580270793,
      "grad_norm": 17.96258544921875,
      "kl": 0.208984375,
      "learning_rate": 4.177949709864603e-07,
      "loss": 0.0084,
      "reward": 1.539582073688507,
      "reward_std": 0.10280460864305496,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5552070736885071,
      "step": 1806
    },
    {
      "completion_length": 108.265625,
      "epoch": 1.1650548033526757,
      "grad_norm": 21.98755645751953,
      "kl": 0.15234375,
      "learning_rate": 4.174725983236622e-07,
      "loss": 0.0061,
      "reward": 1.6323410272598267,
      "reward_std": 0.1224982887506485,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6323409676551819,
      "step": 1807
    },
    {
      "completion_length": 104.3125,
      "epoch": 1.1656995486782722,
      "grad_norm": 19.422637939453125,
      "kl": 0.12890625,
      "learning_rate": 4.171502256608639e-07,
      "loss": 0.0051,
      "reward": 1.560622751712799,
      "reward_std": 0.06334619224071503,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5606227219104767,
      "step": 1808
    },
    {
      "completion_length": 96.25,
      "epoch": 1.1663442940038684,
      "grad_norm": 10.938532829284668,
      "kl": 0.12841796875,
      "learning_rate": 4.168278529980657e-07,
      "loss": 0.0051,
      "reward": 1.5526504516601562,
      "reward_std": 0.09076980128884315,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5526504814624786,
      "step": 1809
    },
    {
      "completion_length": 104.625,
      "epoch": 1.1669890393294649,
      "grad_norm": 21.19317054748535,
      "kl": 0.146484375,
      "learning_rate": 4.165054803352676e-07,
      "loss": 0.0058,
      "reward": 1.4290502071380615,
      "reward_std": 0.1222308985888958,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4290502667427063,
      "step": 1810
    },
    {
      "completion_length": 108.125,
      "epoch": 1.1676337846550613,
      "grad_norm": 10.554473876953125,
      "kl": 0.13671875,
      "learning_rate": 4.161831076724693e-07,
      "loss": 0.0055,
      "reward": 1.443765640258789,
      "reward_std": 0.053497713059186935,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44376565515995026,
      "step": 1811
    },
    {
      "completion_length": 118.703125,
      "epoch": 1.1682785299806577,
      "grad_norm": 23.218820571899414,
      "kl": 0.13037109375,
      "learning_rate": 4.158607350096712e-07,
      "loss": 0.0052,
      "reward": 1.5981783866882324,
      "reward_std": 0.05331315100193024,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5981782972812653,
      "step": 1812
    },
    {
      "completion_length": 126.234375,
      "epoch": 1.168923275306254,
      "grad_norm": 6.66879415512085,
      "kl": 0.130615234375,
      "learning_rate": 4.15538362346873e-07,
      "loss": 0.0052,
      "reward": 1.564515233039856,
      "reward_std": 0.09983974322676659,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.564515233039856,
      "step": 1813
    },
    {
      "completion_length": 112.984375,
      "epoch": 1.1695680206318504,
      "grad_norm": 9.480602264404297,
      "kl": 0.14208984375,
      "learning_rate": 4.152159896840748e-07,
      "loss": 0.0057,
      "reward": 1.7875486016273499,
      "reward_std": 0.09949357807636261,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7875485420227051,
      "step": 1814
    },
    {
      "completion_length": 104.921875,
      "epoch": 1.1702127659574468,
      "grad_norm": 6.718891620635986,
      "kl": 0.11669921875,
      "learning_rate": 4.148936170212766e-07,
      "loss": 0.0047,
      "reward": 1.604787290096283,
      "reward_std": 0.09039689414203167,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6047872602939606,
      "step": 1815
    },
    {
      "completion_length": 113.359375,
      "epoch": 1.1708575112830433,
      "grad_norm": 10.286612510681152,
      "kl": 0.154052734375,
      "learning_rate": 4.145712443584784e-07,
      "loss": 0.0062,
      "reward": 1.5428114533424377,
      "reward_std": 0.06623676046729088,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5428114533424377,
      "step": 1816
    },
    {
      "completion_length": 107.125,
      "epoch": 1.1715022566086395,
      "grad_norm": 18.38083267211914,
      "kl": 0.13232421875,
      "learning_rate": 4.142488716956802e-07,
      "loss": 0.0053,
      "reward": 1.596979796886444,
      "reward_std": 0.1776793971657753,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6126048564910889,
      "step": 1817
    },
    {
      "completion_length": 128.296875,
      "epoch": 1.172147001934236,
      "grad_norm": 12.867851257324219,
      "kl": 0.11962890625,
      "learning_rate": 4.1392649903288204e-07,
      "loss": 0.0048,
      "reward": 1.6555660367012024,
      "reward_std": 0.08787687122821808,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6555660367012024,
      "step": 1818
    },
    {
      "completion_length": 109.84375,
      "epoch": 1.1727917472598324,
      "grad_norm": 77.79759216308594,
      "kl": 0.111328125,
      "learning_rate": 4.136041263700838e-07,
      "loss": 0.0045,
      "reward": 1.6702940464019775,
      "reward_std": 0.08833744376897812,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6702939569950104,
      "step": 1819
    },
    {
      "completion_length": 112.0625,
      "epoch": 1.1734364925854288,
      "grad_norm": 24.407093048095703,
      "kl": 0.125732421875,
      "learning_rate": 4.1328175370728565e-07,
      "loss": 0.005,
      "reward": 1.694697380065918,
      "reward_std": 0.059025075286626816,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.694697380065918,
      "step": 1820
    },
    {
      "completion_length": 119.015625,
      "epoch": 1.1740812379110253,
      "grad_norm": 10.360937118530273,
      "kl": 0.13623046875,
      "learning_rate": 4.1295938104448743e-07,
      "loss": 0.0054,
      "reward": 1.6886966824531555,
      "reward_std": 0.11815246567130089,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6886966526508331,
      "step": 1821
    },
    {
      "completion_length": 121.21875,
      "epoch": 1.1747259832366215,
      "grad_norm": 24.1602783203125,
      "kl": 0.10791015625,
      "learning_rate": 4.126370083816892e-07,
      "loss": 0.0043,
      "reward": 1.549957811832428,
      "reward_std": 0.1373315379023552,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5812077820301056,
      "step": 1822
    },
    {
      "completion_length": 113.140625,
      "epoch": 1.175370728562218,
      "grad_norm": 16.046770095825195,
      "kl": 0.114990234375,
      "learning_rate": 4.1231463571889105e-07,
      "loss": 0.0046,
      "reward": 1.5942580699920654,
      "reward_std": 0.09112752042710781,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5942580997943878,
      "step": 1823
    },
    {
      "completion_length": 114.390625,
      "epoch": 1.1760154738878144,
      "grad_norm": 13.49719524383545,
      "kl": 0.138671875,
      "learning_rate": 4.1199226305609283e-07,
      "loss": 0.0056,
      "reward": 1.601550281047821,
      "reward_std": 0.0931050106883049,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.601550281047821,
      "step": 1824
    },
    {
      "completion_length": 108.75,
      "epoch": 1.1766602192134108,
      "grad_norm": 12.032121658325195,
      "kl": 0.11376953125,
      "learning_rate": 4.1166989039329466e-07,
      "loss": 0.0046,
      "reward": 1.6884644031524658,
      "reward_std": 0.14345430955290794,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7040894031524658,
      "step": 1825
    },
    {
      "completion_length": 110.671875,
      "epoch": 1.177304964539007,
      "grad_norm": 34.89378356933594,
      "kl": 0.11279296875,
      "learning_rate": 4.1134751773049644e-07,
      "loss": 0.0045,
      "reward": 1.675636649131775,
      "reward_std": 0.19111549109220505,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6912615597248077,
      "step": 1826
    },
    {
      "completion_length": 116.640625,
      "epoch": 1.1779497098646035,
      "grad_norm": 14.025679588317871,
      "kl": 0.124267578125,
      "learning_rate": 4.110251450676983e-07,
      "loss": 0.005,
      "reward": 1.477051854133606,
      "reward_std": 0.09772965125739574,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.47705191373825073,
      "step": 1827
    },
    {
      "completion_length": 111.4375,
      "epoch": 1.1785944551902,
      "grad_norm": 12.973898887634277,
      "kl": 0.19384765625,
      "learning_rate": 4.1070277240490005e-07,
      "loss": 0.0077,
      "reward": 1.724359393119812,
      "reward_std": 0.13874923437833786,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7243594825267792,
      "step": 1828
    },
    {
      "completion_length": 129.140625,
      "epoch": 1.1792392005157963,
      "grad_norm": 14.122653007507324,
      "kl": 0.121337890625,
      "learning_rate": 4.103803997421019e-07,
      "loss": 0.0048,
      "reward": 1.600561499595642,
      "reward_std": 0.10613074153661728,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6005614995956421,
      "step": 1829
    },
    {
      "completion_length": 108.09375,
      "epoch": 1.1798839458413926,
      "grad_norm": 18.514957427978516,
      "kl": 0.153564453125,
      "learning_rate": 4.1005802707930367e-07,
      "loss": 0.0061,
      "reward": 1.6020503640174866,
      "reward_std": 0.08536173682659864,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.602050393819809,
      "step": 1830
    },
    {
      "completion_length": 106.703125,
      "epoch": 1.180528691166989,
      "grad_norm": 12.282980918884277,
      "kl": 0.11474609375,
      "learning_rate": 4.097356544165055e-07,
      "loss": 0.0046,
      "reward": 1.552680253982544,
      "reward_std": 0.05671044811606407,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5526802390813828,
      "step": 1831
    },
    {
      "completion_length": 118.015625,
      "epoch": 1.1811734364925854,
      "grad_norm": 12.24289321899414,
      "kl": 0.1318359375,
      "learning_rate": 4.094132817537073e-07,
      "loss": 0.0053,
      "reward": 1.5175416469573975,
      "reward_std": 0.07208650559186935,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5175416022539139,
      "step": 1832
    },
    {
      "completion_length": 107.796875,
      "epoch": 1.1818181818181819,
      "grad_norm": 10.43206787109375,
      "kl": 0.142578125,
      "learning_rate": 4.090909090909091e-07,
      "loss": 0.0057,
      "reward": 1.4518251419067383,
      "reward_std": 0.07507746666669846,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.45182517170906067,
      "step": 1833
    },
    {
      "completion_length": 113.875,
      "epoch": 1.1824629271437783,
      "grad_norm": 30.288999557495117,
      "kl": 0.1083984375,
      "learning_rate": 4.087685364281109e-07,
      "loss": 0.0043,
      "reward": 1.7571146488189697,
      "reward_std": 0.07010109722614288,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7571147084236145,
      "step": 1834
    },
    {
      "completion_length": 112.453125,
      "epoch": 1.1831076724693745,
      "grad_norm": 12.52745532989502,
      "kl": 0.215087890625,
      "learning_rate": 4.084461637653127e-07,
      "loss": 0.0086,
      "reward": 1.6558008193969727,
      "reward_std": 0.10477161407470703,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6558008193969727,
      "step": 1835
    },
    {
      "completion_length": 109.578125,
      "epoch": 1.183752417794971,
      "grad_norm": 13.707616806030273,
      "kl": 0.116455078125,
      "learning_rate": 4.081237911025145e-07,
      "loss": 0.0047,
      "reward": 1.6503602862358093,
      "reward_std": 0.08982540853321552,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6503602713346481,
      "step": 1836
    },
    {
      "completion_length": 105.71875,
      "epoch": 1.1843971631205674,
      "grad_norm": 11.990213394165039,
      "kl": 0.11669921875,
      "learning_rate": 4.078014184397163e-07,
      "loss": 0.0047,
      "reward": 1.503011703491211,
      "reward_std": 0.22921422868967056,
      "rewards/format_reward": 0.96875,
      "rewards/iou_timestamp_reward": 0.5342617332935333,
      "step": 1837
    },
    {
      "completion_length": 108.46875,
      "epoch": 1.1850419084461639,
      "grad_norm": 20.66301727294922,
      "kl": 0.12109375,
      "learning_rate": 4.074790457769181e-07,
      "loss": 0.0048,
      "reward": 1.5639979243278503,
      "reward_std": 0.11984455585479736,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5639979243278503,
      "step": 1838
    },
    {
      "completion_length": 126.203125,
      "epoch": 1.18568665377176,
      "grad_norm": 63.44125747680664,
      "kl": 0.101806640625,
      "learning_rate": 4.071566731141199e-07,
      "loss": 0.0041,
      "reward": 1.6595558524131775,
      "reward_std": 0.11242682486772537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6595558822154999,
      "step": 1839
    },
    {
      "completion_length": 110.40625,
      "epoch": 1.1863313990973565,
      "grad_norm": 19.61737632751465,
      "kl": 0.1181640625,
      "learning_rate": 4.0683430045132174e-07,
      "loss": 0.0047,
      "reward": 1.5931558012962341,
      "reward_std": 0.14882973581552505,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5931558012962341,
      "step": 1840
    },
    {
      "completion_length": 114.671875,
      "epoch": 1.186976144422953,
      "grad_norm": 27.75486183166504,
      "kl": 0.1435546875,
      "learning_rate": 4.065119277885235e-07,
      "loss": 0.0057,
      "reward": 1.7199392914772034,
      "reward_std": 0.12808328121900558,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.719939261674881,
      "step": 1841
    },
    {
      "completion_length": 109.953125,
      "epoch": 1.1876208897485494,
      "grad_norm": 13.68565845489502,
      "kl": 0.10498046875,
      "learning_rate": 4.0618955512572535e-07,
      "loss": 0.0042,
      "reward": 1.5208486318588257,
      "reward_std": 0.14144017547369003,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5364736467599869,
      "step": 1842
    },
    {
      "completion_length": 113.890625,
      "epoch": 1.1882656350741456,
      "grad_norm": 15.589136123657227,
      "kl": 0.11279296875,
      "learning_rate": 4.0586718246292713e-07,
      "loss": 0.0045,
      "reward": 1.7398034930229187,
      "reward_std": 0.04764362797141075,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7398034334182739,
      "step": 1843
    },
    {
      "completion_length": 105.359375,
      "epoch": 1.188910380399742,
      "grad_norm": 14.390912055969238,
      "kl": 0.13037109375,
      "learning_rate": 4.0554480980012897e-07,
      "loss": 0.0052,
      "reward": 1.3299148678779602,
      "reward_std": 0.09040863066911697,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3299148678779602,
      "step": 1844
    },
    {
      "completion_length": 107.78125,
      "epoch": 1.1895551257253385,
      "grad_norm": 28.451305389404297,
      "kl": 0.135009765625,
      "learning_rate": 4.0522243713733075e-07,
      "loss": 0.0054,
      "reward": 1.469842255115509,
      "reward_std": 0.11970901861786842,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4698422998189926,
      "step": 1845
    },
    {
      "completion_length": 107.296875,
      "epoch": 1.190199871050935,
      "grad_norm": 13.97714900970459,
      "kl": 0.117919921875,
      "learning_rate": 4.049000644745326e-07,
      "loss": 0.0047,
      "reward": 1.7399972081184387,
      "reward_std": 0.11769672483205795,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7399971783161163,
      "step": 1846
    },
    {
      "completion_length": 113.75,
      "epoch": 1.1908446163765314,
      "grad_norm": 8.553421020507812,
      "kl": 0.119140625,
      "learning_rate": 4.0457769181173436e-07,
      "loss": 0.0048,
      "reward": 1.6283788084983826,
      "reward_std": 0.11559978127479553,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6283787786960602,
      "step": 1847
    },
    {
      "completion_length": 104.359375,
      "epoch": 1.1914893617021276,
      "grad_norm": 8.622238159179688,
      "kl": 0.14990234375,
      "learning_rate": 4.0425531914893614e-07,
      "loss": 0.006,
      "reward": 1.5790454745292664,
      "reward_std": 0.09102013148367405,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5790454745292664,
      "step": 1848
    },
    {
      "completion_length": 103.78125,
      "epoch": 1.192134107027724,
      "grad_norm": 19.665542602539062,
      "kl": 0.1298828125,
      "learning_rate": 4.03932946486138e-07,
      "loss": 0.0052,
      "reward": 1.6341629028320312,
      "reward_std": 0.07100488245487213,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6341629028320312,
      "step": 1849
    },
    {
      "completion_length": 109.203125,
      "epoch": 1.1927788523533205,
      "grad_norm": 23.64558219909668,
      "kl": 0.109619140625,
      "learning_rate": 4.0361057382333975e-07,
      "loss": 0.0044,
      "reward": 1.6545790433883667,
      "reward_std": 0.0724497102200985,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6545790731906891,
      "step": 1850
    },
    {
      "completion_length": 109.15625,
      "epoch": 1.1934235976789167,
      "grad_norm": 10.852958679199219,
      "kl": 0.12451171875,
      "learning_rate": 4.032882011605416e-07,
      "loss": 0.005,
      "reward": 1.6551001071929932,
      "reward_std": 0.10190777480602264,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6551000475883484,
      "step": 1851
    },
    {
      "completion_length": 107.171875,
      "epoch": 1.1940683430045131,
      "grad_norm": 7.953454494476318,
      "kl": 0.130859375,
      "learning_rate": 4.0296582849774337e-07,
      "loss": 0.0053,
      "reward": 1.5733691453933716,
      "reward_std": 0.09967935457825661,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.573369175195694,
      "step": 1852
    },
    {
      "completion_length": 114.265625,
      "epoch": 1.1947130883301096,
      "grad_norm": 11.328710556030273,
      "kl": 0.14306640625,
      "learning_rate": 4.026434558349452e-07,
      "loss": 0.0057,
      "reward": 1.7428075671195984,
      "reward_std": 0.10012370720505714,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7428075969219208,
      "step": 1853
    },
    {
      "completion_length": 123.390625,
      "epoch": 1.195357833655706,
      "grad_norm": 20.28865623474121,
      "kl": 0.103759765625,
      "learning_rate": 4.02321083172147e-07,
      "loss": 0.0041,
      "reward": 1.5555278062820435,
      "reward_std": 0.09157108515501022,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5555277466773987,
      "step": 1854
    },
    {
      "completion_length": 122.921875,
      "epoch": 1.1960025789813025,
      "grad_norm": 8.482815742492676,
      "kl": 0.11962890625,
      "learning_rate": 4.019987105093488e-07,
      "loss": 0.0048,
      "reward": 1.5178706645965576,
      "reward_std": 0.08336883038282394,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5178706347942352,
      "step": 1855
    },
    {
      "completion_length": 118.125,
      "epoch": 1.1966473243068987,
      "grad_norm": 9.631011962890625,
      "kl": 0.101318359375,
      "learning_rate": 4.016763378465506e-07,
      "loss": 0.0041,
      "reward": 1.659381926059723,
      "reward_std": 0.06883524730801582,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6593818664550781,
      "step": 1856
    },
    {
      "completion_length": 119.8125,
      "epoch": 1.1972920696324951,
      "grad_norm": 8.929444313049316,
      "kl": 0.103515625,
      "learning_rate": 4.0135396518375243e-07,
      "loss": 0.0041,
      "reward": 1.5685328841209412,
      "reward_std": 0.10261768102645874,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5685328543186188,
      "step": 1857
    },
    {
      "completion_length": 105.3125,
      "epoch": 1.1979368149580916,
      "grad_norm": 32.578330993652344,
      "kl": 0.1083984375,
      "learning_rate": 4.010315925209542e-07,
      "loss": 0.0043,
      "reward": 1.5939223766326904,
      "reward_std": 0.10584653168916702,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5939223766326904,
      "step": 1858
    },
    {
      "completion_length": 119.296875,
      "epoch": 1.198581560283688,
      "grad_norm": 9.094733238220215,
      "kl": 0.10302734375,
      "learning_rate": 4.0070921985815604e-07,
      "loss": 0.0041,
      "reward": 1.7315629124641418,
      "reward_std": 0.10077636688947678,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7315629422664642,
      "step": 1859
    },
    {
      "completion_length": 110.34375,
      "epoch": 1.1992263056092844,
      "grad_norm": 13.449752807617188,
      "kl": 0.118408203125,
      "learning_rate": 4.003868471953578e-07,
      "loss": 0.0047,
      "reward": 1.6524174809455872,
      "reward_std": 0.13661283254623413,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6524175405502319,
      "step": 1860
    },
    {
      "completion_length": 110.25,
      "epoch": 1.1998710509348807,
      "grad_norm": 11.15349292755127,
      "kl": 0.117431640625,
      "learning_rate": 4.0006447453255966e-07,
      "loss": 0.0047,
      "reward": 1.5107752680778503,
      "reward_std": 0.21871937066316605,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5264002084732056,
      "step": 1861
    },
    {
      "completion_length": 126.234375,
      "epoch": 1.200515796260477,
      "grad_norm": 8.474930763244629,
      "kl": 0.1123046875,
      "learning_rate": 3.9974210186976144e-07,
      "loss": 0.0045,
      "reward": 1.515737771987915,
      "reward_std": 0.11956487596035004,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.515737771987915,
      "step": 1862
    },
    {
      "completion_length": 118.875,
      "epoch": 1.2011605415860735,
      "grad_norm": 14.834139823913574,
      "kl": 0.12109375,
      "learning_rate": 3.994197292069632e-07,
      "loss": 0.0048,
      "reward": 1.6979880928993225,
      "reward_std": 0.17543792724609375,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.7136130630970001,
      "step": 1863
    },
    {
      "completion_length": 106.984375,
      "epoch": 1.2018052869116698,
      "grad_norm": 18.74297332763672,
      "kl": 0.104736328125,
      "learning_rate": 3.9909735654416505e-07,
      "loss": 0.0042,
      "reward": 1.5749844312667847,
      "reward_std": 0.13023612275719643,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5906094610691071,
      "step": 1864
    },
    {
      "completion_length": 109.015625,
      "epoch": 1.2024500322372662,
      "grad_norm": 10.109322547912598,
      "kl": 0.14599609375,
      "learning_rate": 3.9877498388136683e-07,
      "loss": 0.0058,
      "reward": 1.6240103244781494,
      "reward_std": 0.18006663024425507,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6240103244781494,
      "step": 1865
    },
    {
      "completion_length": 115.53125,
      "epoch": 1.2030947775628626,
      "grad_norm": 98.04276275634766,
      "kl": 0.119384765625,
      "learning_rate": 3.9845261121856867e-07,
      "loss": 0.0048,
      "reward": 1.6898424625396729,
      "reward_std": 0.08776196092367172,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6898424625396729,
      "step": 1866
    },
    {
      "completion_length": 111.0625,
      "epoch": 1.203739522888459,
      "grad_norm": 16.498126983642578,
      "kl": 0.114990234375,
      "learning_rate": 3.9813023855577045e-07,
      "loss": 0.0046,
      "reward": 1.4743621349334717,
      "reward_std": 0.20478830486536026,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.4899871349334717,
      "step": 1867
    },
    {
      "completion_length": 119.71875,
      "epoch": 1.2043842682140555,
      "grad_norm": 15.135804176330566,
      "kl": 0.11572265625,
      "learning_rate": 3.978078658929723e-07,
      "loss": 0.0046,
      "reward": 1.4399540424346924,
      "reward_std": 0.11498159915208817,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.43995407223701477,
      "step": 1868
    },
    {
      "completion_length": 111.0,
      "epoch": 1.2050290135396517,
      "grad_norm": 10.168272972106934,
      "kl": 0.118408203125,
      "learning_rate": 3.9748549323017406e-07,
      "loss": 0.0047,
      "reward": 1.5744973421096802,
      "reward_std": 0.05219046585261822,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5744973719120026,
      "step": 1869
    },
    {
      "completion_length": 117.65625,
      "epoch": 1.2056737588652482,
      "grad_norm": 7.834216117858887,
      "kl": 0.12060546875,
      "learning_rate": 3.971631205673759e-07,
      "loss": 0.0048,
      "reward": 1.6258432269096375,
      "reward_std": 0.07550669834017754,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6258432269096375,
      "step": 1870
    },
    {
      "completion_length": 98.6875,
      "epoch": 1.2063185041908446,
      "grad_norm": 13.367279052734375,
      "kl": 0.12451171875,
      "learning_rate": 3.968407479045777e-07,
      "loss": 0.005,
      "reward": 1.5441078543663025,
      "reward_std": 0.11310616880655289,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5441079139709473,
      "step": 1871
    },
    {
      "completion_length": 114.359375,
      "epoch": 1.206963249516441,
      "grad_norm": 11.95669937133789,
      "kl": 0.113037109375,
      "learning_rate": 3.965183752417795e-07,
      "loss": 0.0045,
      "reward": 1.6491506695747375,
      "reward_std": 0.15457647293806076,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6647757291793823,
      "step": 1872
    },
    {
      "completion_length": 123.53125,
      "epoch": 1.2076079948420375,
      "grad_norm": 23.579849243164062,
      "kl": 0.103515625,
      "learning_rate": 3.961960025789813e-07,
      "loss": 0.0041,
      "reward": 1.4970921874046326,
      "reward_std": 0.1336561180651188,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.497092142701149,
      "step": 1873
    },
    {
      "completion_length": 106.109375,
      "epoch": 1.2082527401676337,
      "grad_norm": 12.425701141357422,
      "kl": 0.1162109375,
      "learning_rate": 3.958736299161831e-07,
      "loss": 0.0046,
      "reward": 1.50269216299057,
      "reward_std": 0.0642969198524952,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5026921331882477,
      "step": 1874
    },
    {
      "completion_length": 100.109375,
      "epoch": 1.2088974854932302,
      "grad_norm": 9.212443351745605,
      "kl": 0.1416015625,
      "learning_rate": 3.955512572533849e-07,
      "loss": 0.0057,
      "reward": 1.757175624370575,
      "reward_std": 0.17486996948719025,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.772800624370575,
      "step": 1875
    },
    {
      "completion_length": 112.890625,
      "epoch": 1.2095422308188266,
      "grad_norm": 12.235011100769043,
      "kl": 0.1220703125,
      "learning_rate": 3.952288845905867e-07,
      "loss": 0.0049,
      "reward": 1.5499593615531921,
      "reward_std": 0.12618184834718704,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5499593615531921,
      "step": 1876
    },
    {
      "completion_length": 113.53125,
      "epoch": 1.2101869761444228,
      "grad_norm": 41.290550231933594,
      "kl": 0.1318359375,
      "learning_rate": 3.949065119277885e-07,
      "loss": 0.0053,
      "reward": 1.5221721529960632,
      "reward_std": 0.10043122246861458,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.522172138094902,
      "step": 1877
    },
    {
      "completion_length": 116.4375,
      "epoch": 1.2108317214700193,
      "grad_norm": 10.30317497253418,
      "kl": 0.13720703125,
      "learning_rate": 3.945841392649903e-07,
      "loss": 0.0055,
      "reward": 1.652651309967041,
      "reward_std": 0.14301005005836487,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6682763397693634,
      "step": 1878
    },
    {
      "completion_length": 120.375,
      "epoch": 1.2114764667956157,
      "grad_norm": 7.7513580322265625,
      "kl": 0.140625,
      "learning_rate": 3.9426176660219213e-07,
      "loss": 0.0056,
      "reward": 1.447944164276123,
      "reward_std": 0.09541654214262962,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.44794413447380066,
      "step": 1879
    },
    {
      "completion_length": 119.65625,
      "epoch": 1.2121212121212122,
      "grad_norm": 16.127681732177734,
      "kl": 0.120361328125,
      "learning_rate": 3.939393939393939e-07,
      "loss": 0.0048,
      "reward": 1.594392716884613,
      "reward_std": 0.08314290270209312,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5943926572799683,
      "step": 1880
    },
    {
      "completion_length": 128.78125,
      "epoch": 1.2127659574468086,
      "grad_norm": 56.94392013549805,
      "kl": 0.111328125,
      "learning_rate": 3.9361702127659574e-07,
      "loss": 0.0045,
      "reward": 1.7492846846580505,
      "reward_std": 0.07751237228512764,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7492846846580505,
      "step": 1881
    },
    {
      "completion_length": 115.03125,
      "epoch": 1.2134107027724048,
      "grad_norm": 22.434030532836914,
      "kl": 0.15625,
      "learning_rate": 3.932946486137975e-07,
      "loss": 0.0063,
      "reward": 1.5479236245155334,
      "reward_std": 0.08889007940888405,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5479235649108887,
      "step": 1882
    },
    {
      "completion_length": 108.109375,
      "epoch": 1.2140554480980013,
      "grad_norm": 7.91255521774292,
      "kl": 0.115234375,
      "learning_rate": 3.9297227595099936e-07,
      "loss": 0.0046,
      "reward": 1.674900233745575,
      "reward_std": 0.08419017121195793,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6749002635478973,
      "step": 1883
    },
    {
      "completion_length": 120.5625,
      "epoch": 1.2147001934235977,
      "grad_norm": 42.26613998413086,
      "kl": 0.12255859375,
      "learning_rate": 3.9264990328820114e-07,
      "loss": 0.0049,
      "reward": 1.6492977738380432,
      "reward_std": 0.14469093456864357,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6492977738380432,
      "step": 1884
    },
    {
      "completion_length": 124.046875,
      "epoch": 1.2153449387491941,
      "grad_norm": 35.12214279174805,
      "kl": 0.1220703125,
      "learning_rate": 3.9232753062540297e-07,
      "loss": 0.0049,
      "reward": 1.5660077929496765,
      "reward_std": 0.08853030204772949,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5660077631473541,
      "step": 1885
    },
    {
      "completion_length": 119.0,
      "epoch": 1.2159896840747906,
      "grad_norm": 10.311363220214844,
      "kl": 0.111083984375,
      "learning_rate": 3.9200515796260475e-07,
      "loss": 0.0044,
      "reward": 1.7127981781959534,
      "reward_std": 0.1427604779601097,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.712798148393631,
      "step": 1886
    },
    {
      "completion_length": 117.796875,
      "epoch": 1.2166344294003868,
      "grad_norm": 10.61994457244873,
      "kl": 0.1171875,
      "learning_rate": 3.916827852998066e-07,
      "loss": 0.0047,
      "reward": 1.6093450784683228,
      "reward_std": 0.1268884353339672,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6093450784683228,
      "step": 1887
    },
    {
      "completion_length": 125.671875,
      "epoch": 1.2172791747259832,
      "grad_norm": 10.94974136352539,
      "kl": 0.108642578125,
      "learning_rate": 3.9136041263700837e-07,
      "loss": 0.0043,
      "reward": 1.7339577674865723,
      "reward_std": 0.06080514006316662,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7339577078819275,
      "step": 1888
    },
    {
      "completion_length": 130.171875,
      "epoch": 1.2179239200515797,
      "grad_norm": 29.69418716430664,
      "kl": 0.11669921875,
      "learning_rate": 3.9103803997421015e-07,
      "loss": 0.0047,
      "reward": 1.546896517276764,
      "reward_std": 0.10660860687494278,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5468964874744415,
      "step": 1889
    },
    {
      "completion_length": 109.1875,
      "epoch": 1.218568665377176,
      "grad_norm": 10.2553071975708,
      "kl": 0.129638671875,
      "learning_rate": 3.90715667311412e-07,
      "loss": 0.0052,
      "reward": 1.7039321660995483,
      "reward_std": 0.1126185953617096,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7039321660995483,
      "step": 1890
    },
    {
      "completion_length": 127.015625,
      "epoch": 1.2192134107027723,
      "grad_norm": 15.430004119873047,
      "kl": 0.09765625,
      "learning_rate": 3.9039329464861376e-07,
      "loss": 0.0039,
      "reward": 1.644881248474121,
      "reward_std": 0.0988176241517067,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6448812484741211,
      "step": 1891
    },
    {
      "completion_length": 120.84375,
      "epoch": 1.2198581560283688,
      "grad_norm": 52.64883041381836,
      "kl": 0.114013671875,
      "learning_rate": 3.900709219858156e-07,
      "loss": 0.0046,
      "reward": 1.7061986923217773,
      "reward_std": 0.0526033453643322,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7061986327171326,
      "step": 1892
    },
    {
      "completion_length": 136.9375,
      "epoch": 1.2205029013539652,
      "grad_norm": 11.081215858459473,
      "kl": 0.13623046875,
      "learning_rate": 3.897485493230174e-07,
      "loss": 0.0054,
      "reward": 1.4815837144851685,
      "reward_std": 0.07057693414390087,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48158374428749084,
      "step": 1893
    },
    {
      "completion_length": 113.53125,
      "epoch": 1.2211476466795617,
      "grad_norm": 9.108400344848633,
      "kl": 0.13623046875,
      "learning_rate": 3.894261766602192e-07,
      "loss": 0.0054,
      "reward": 1.6832290291786194,
      "reward_std": 0.08856482990086079,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6988540291786194,
      "step": 1894
    },
    {
      "completion_length": 116.5,
      "epoch": 1.2217923920051579,
      "grad_norm": 33.61598587036133,
      "kl": 0.107666015625,
      "learning_rate": 3.89103803997421e-07,
      "loss": 0.0043,
      "reward": 1.6324860453605652,
      "reward_std": 0.09398233145475388,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6324860155582428,
      "step": 1895
    },
    {
      "completion_length": 117.90625,
      "epoch": 1.2224371373307543,
      "grad_norm": 14.763751029968262,
      "kl": 0.127685546875,
      "learning_rate": 3.887814313346228e-07,
      "loss": 0.0051,
      "reward": 1.5660215020179749,
      "reward_std": 0.09480106830596924,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5660214722156525,
      "step": 1896
    },
    {
      "completion_length": 119.46875,
      "epoch": 1.2230818826563508,
      "grad_norm": 23.6367244720459,
      "kl": 0.1298828125,
      "learning_rate": 3.884590586718246e-07,
      "loss": 0.0052,
      "reward": 1.598044991493225,
      "reward_std": 0.121733158826828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5980449616909027,
      "step": 1897
    },
    {
      "completion_length": 109.390625,
      "epoch": 1.2237266279819472,
      "grad_norm": 12.922637939453125,
      "kl": 0.1123046875,
      "learning_rate": 3.8813668600902644e-07,
      "loss": 0.0045,
      "reward": 1.6700684428215027,
      "reward_std": 0.07409593462944031,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6700685024261475,
      "step": 1898
    },
    {
      "completion_length": 119.078125,
      "epoch": 1.2243713733075436,
      "grad_norm": 42.0638542175293,
      "kl": 0.11474609375,
      "learning_rate": 3.878143133462282e-07,
      "loss": 0.0046,
      "reward": 1.6279435753822327,
      "reward_std": 0.11568474024534225,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6279435753822327,
      "step": 1899
    },
    {
      "completion_length": 120.109375,
      "epoch": 1.2250161186331399,
      "grad_norm": 7.401854038238525,
      "kl": 0.1142578125,
      "learning_rate": 3.8749194068343005e-07,
      "loss": 0.0046,
      "reward": 1.5463709831237793,
      "reward_std": 0.12193901836872101,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5463710427284241,
      "step": 1900
    },
    {
      "completion_length": 119.0625,
      "epoch": 1.2256608639587363,
      "grad_norm": 10.23751163482666,
      "kl": 0.107421875,
      "learning_rate": 3.8716956802063183e-07,
      "loss": 0.0043,
      "reward": 1.59816575050354,
      "reward_std": 0.08526627346873283,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5981658101081848,
      "step": 1901
    },
    {
      "completion_length": 128.453125,
      "epoch": 1.2263056092843327,
      "grad_norm": 8.265456199645996,
      "kl": 0.098876953125,
      "learning_rate": 3.868471953578336e-07,
      "loss": 0.004,
      "reward": 1.820991039276123,
      "reward_std": 0.05691416189074516,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.8209910690784454,
      "step": 1902
    },
    {
      "completion_length": 116.3125,
      "epoch": 1.226950354609929,
      "grad_norm": 10.897770881652832,
      "kl": 0.104736328125,
      "learning_rate": 3.8652482269503544e-07,
      "loss": 0.0042,
      "reward": 1.617471992969513,
      "reward_std": 0.08763501048088074,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6174720227718353,
      "step": 1903
    },
    {
      "completion_length": 117.328125,
      "epoch": 1.2275950999355254,
      "grad_norm": 11.371556282043457,
      "kl": 0.129150390625,
      "learning_rate": 3.862024500322372e-07,
      "loss": 0.0052,
      "reward": 1.4868090152740479,
      "reward_std": 0.06387560069561005,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48680898547172546,
      "step": 1904
    },
    {
      "completion_length": 127.53125,
      "epoch": 1.2282398452611218,
      "grad_norm": 7.77946138381958,
      "kl": 0.12060546875,
      "learning_rate": 3.8588007736943906e-07,
      "loss": 0.0048,
      "reward": 1.6102416515350342,
      "reward_std": 0.15693595260381699,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6102416068315506,
      "step": 1905
    },
    {
      "completion_length": 111.6875,
      "epoch": 1.2288845905867183,
      "grad_norm": 16.03278160095215,
      "kl": 0.107666015625,
      "learning_rate": 3.8555770470664084e-07,
      "loss": 0.0043,
      "reward": 1.5970102548599243,
      "reward_std": 0.06430243700742722,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5970102250576019,
      "step": 1906
    },
    {
      "completion_length": 119.453125,
      "epoch": 1.2295293359123147,
      "grad_norm": 10.253130912780762,
      "kl": 0.14599609375,
      "learning_rate": 3.8523533204384267e-07,
      "loss": 0.0058,
      "reward": 1.6699411273002625,
      "reward_std": 0.09906578063964844,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6699410974979401,
      "step": 1907
    },
    {
      "completion_length": 120.21875,
      "epoch": 1.230174081237911,
      "grad_norm": 10.90766429901123,
      "kl": 0.116943359375,
      "learning_rate": 3.8491295938104445e-07,
      "loss": 0.0047,
      "reward": 1.7290301322937012,
      "reward_std": 0.12131063267588615,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7290300726890564,
      "step": 1908
    },
    {
      "completion_length": 129.828125,
      "epoch": 1.2308188265635074,
      "grad_norm": 9.48066234588623,
      "kl": 0.1123046875,
      "learning_rate": 3.845905867182463e-07,
      "loss": 0.0045,
      "reward": 1.6760023832321167,
      "reward_std": 0.14010852575302124,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6916274130344391,
      "step": 1909
    },
    {
      "completion_length": 120.15625,
      "epoch": 1.2314635718891038,
      "grad_norm": 13.372663497924805,
      "kl": 0.106201171875,
      "learning_rate": 3.8426821405544807e-07,
      "loss": 0.0043,
      "reward": 1.489621639251709,
      "reward_std": 0.0755012072622776,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48962169885635376,
      "step": 1910
    },
    {
      "completion_length": 136.375,
      "epoch": 1.2321083172147003,
      "grad_norm": 32.22721481323242,
      "kl": 0.126708984375,
      "learning_rate": 3.839458413926499e-07,
      "loss": 0.0051,
      "reward": 1.5128714442253113,
      "reward_std": 0.15325216948986053,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5284964740276337,
      "step": 1911
    },
    {
      "completion_length": 114.5625,
      "epoch": 1.2327530625402967,
      "grad_norm": 15.361464500427246,
      "kl": 0.11865234375,
      "learning_rate": 3.836234687298517e-07,
      "loss": 0.0047,
      "reward": 1.4920198321342468,
      "reward_std": 0.10974390432238579,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4920198917388916,
      "step": 1912
    },
    {
      "completion_length": 126.1875,
      "epoch": 1.233397807865893,
      "grad_norm": 18.84747886657715,
      "kl": 0.112548828125,
      "learning_rate": 3.833010960670535e-07,
      "loss": 0.0045,
      "reward": 1.4142240285873413,
      "reward_std": 0.04326391592621803,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4142240881919861,
      "step": 1913
    },
    {
      "completion_length": 126.609375,
      "epoch": 1.2340425531914894,
      "grad_norm": 25.70592498779297,
      "kl": 0.120361328125,
      "learning_rate": 3.829787234042553e-07,
      "loss": 0.0048,
      "reward": 1.6135888695716858,
      "reward_std": 0.16180770099163055,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.613588809967041,
      "step": 1914
    },
    {
      "completion_length": 125.203125,
      "epoch": 1.2346872985170858,
      "grad_norm": 30.81764793395996,
      "kl": 0.122802734375,
      "learning_rate": 3.8265635074145713e-07,
      "loss": 0.0049,
      "reward": 1.570852518081665,
      "reward_std": 0.12249016016721725,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5708525478839874,
      "step": 1915
    },
    {
      "completion_length": 126.09375,
      "epoch": 1.235332043842682,
      "grad_norm": 22.804471969604492,
      "kl": 0.116455078125,
      "learning_rate": 3.823339780786589e-07,
      "loss": 0.0047,
      "reward": 1.7204737663269043,
      "reward_std": 0.04909878969192505,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7204737663269043,
      "step": 1916
    },
    {
      "completion_length": 106.25,
      "epoch": 1.2359767891682785,
      "grad_norm": 22.267070770263672,
      "kl": 0.126220703125,
      "learning_rate": 3.820116054158607e-07,
      "loss": 0.0051,
      "reward": 1.7154393792152405,
      "reward_std": 0.1028110608458519,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7154394090175629,
      "step": 1917
    },
    {
      "completion_length": 124.78125,
      "epoch": 1.236621534493875,
      "grad_norm": 29.534210205078125,
      "kl": 0.108642578125,
      "learning_rate": 3.816892327530625e-07,
      "loss": 0.0043,
      "reward": 1.46629136800766,
      "reward_std": 0.09430722147226334,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4662914276123047,
      "step": 1918
    },
    {
      "completion_length": 121.921875,
      "epoch": 1.2372662798194713,
      "grad_norm": 15.665167808532715,
      "kl": 0.10986328125,
      "learning_rate": 3.813668600902643e-07,
      "loss": 0.0044,
      "reward": 1.5596240758895874,
      "reward_std": 0.10030604526400566,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5596240758895874,
      "step": 1919
    },
    {
      "completion_length": 114.5,
      "epoch": 1.2379110251450678,
      "grad_norm": 8.512285232543945,
      "kl": 0.115234375,
      "learning_rate": 3.8104448742746614e-07,
      "loss": 0.0046,
      "reward": 1.6296950578689575,
      "reward_std": 0.05596102774143219,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6296949684619904,
      "step": 1920
    },
    {
      "completion_length": 115.8125,
      "epoch": 1.238555770470664,
      "grad_norm": 79.31735229492188,
      "kl": 0.107666015625,
      "learning_rate": 3.807221147646679e-07,
      "loss": 0.0043,
      "reward": 1.5508543848991394,
      "reward_std": 0.09145446866750717,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.550854355096817,
      "step": 1921
    },
    {
      "completion_length": 119.015625,
      "epoch": 1.2392005157962604,
      "grad_norm": 11.683690071105957,
      "kl": 0.131103515625,
      "learning_rate": 3.8039974210186975e-07,
      "loss": 0.0052,
      "reward": 1.7141978740692139,
      "reward_std": 0.11888259276747704,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7141977846622467,
      "step": 1922
    },
    {
      "completion_length": 117.203125,
      "epoch": 1.239845261121857,
      "grad_norm": 24.96306800842285,
      "kl": 0.123779296875,
      "learning_rate": 3.8007736943907153e-07,
      "loss": 0.0049,
      "reward": 1.646182358264923,
      "reward_std": 0.09359883144497871,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6461824774742126,
      "step": 1923
    },
    {
      "completion_length": 119.828125,
      "epoch": 1.2404900064474533,
      "grad_norm": 13.091527938842773,
      "kl": 0.125,
      "learning_rate": 3.7975499677627336e-07,
      "loss": 0.005,
      "reward": 1.4956902861595154,
      "reward_std": 0.10586478561162949,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4956902712583542,
      "step": 1924
    },
    {
      "completion_length": 116.171875,
      "epoch": 1.2411347517730495,
      "grad_norm": 9.332618713378906,
      "kl": 0.100341796875,
      "learning_rate": 3.7943262411347514e-07,
      "loss": 0.004,
      "reward": 1.715123176574707,
      "reward_std": 0.056105077266693115,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7151232659816742,
      "step": 1925
    },
    {
      "completion_length": 124.453125,
      "epoch": 1.241779497098646,
      "grad_norm": 6.001155853271484,
      "kl": 0.1201171875,
      "learning_rate": 3.79110251450677e-07,
      "loss": 0.0048,
      "reward": 1.5464325547218323,
      "reward_std": 0.09315594844520092,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.546432614326477,
      "step": 1926
    },
    {
      "completion_length": 117.265625,
      "epoch": 1.2424242424242424,
      "grad_norm": 10.31252384185791,
      "kl": 0.105224609375,
      "learning_rate": 3.7878787878787876e-07,
      "loss": 0.0042,
      "reward": 1.5988069772720337,
      "reward_std": 0.0450415788218379,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5988070219755173,
      "step": 1927
    },
    {
      "completion_length": 122.15625,
      "epoch": 1.2430689877498389,
      "grad_norm": 10.713313102722168,
      "kl": 0.1318359375,
      "learning_rate": 3.784655061250806e-07,
      "loss": 0.0053,
      "reward": 1.5424224138259888,
      "reward_std": 0.0438772514462471,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5424223840236664,
      "step": 1928
    },
    {
      "completion_length": 110.421875,
      "epoch": 1.243713733075435,
      "grad_norm": 16.321693420410156,
      "kl": 0.108642578125,
      "learning_rate": 3.7814313346228237e-07,
      "loss": 0.0043,
      "reward": 1.7239617109298706,
      "reward_std": 0.07088684663176537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.723961740732193,
      "step": 1929
    },
    {
      "completion_length": 116.0,
      "epoch": 1.2443584784010315,
      "grad_norm": 12.18964672088623,
      "kl": 0.130859375,
      "learning_rate": 3.7782076079948415e-07,
      "loss": 0.0052,
      "reward": 1.5831393599510193,
      "reward_std": 0.1350330039858818,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5831393897533417,
      "step": 1930
    },
    {
      "completion_length": 107.234375,
      "epoch": 1.245003223726628,
      "grad_norm": 26.728723526000977,
      "kl": 0.1142578125,
      "learning_rate": 3.77498388136686e-07,
      "loss": 0.0046,
      "reward": 1.5077446699142456,
      "reward_std": 0.1127016618847847,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5077447444200516,
      "step": 1931
    },
    {
      "completion_length": 114.515625,
      "epoch": 1.2456479690522244,
      "grad_norm": 14.777571678161621,
      "kl": 0.11279296875,
      "learning_rate": 3.7717601547388777e-07,
      "loss": 0.0045,
      "reward": 1.5410062074661255,
      "reward_std": 0.09689564630389214,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5410062372684479,
      "step": 1932
    },
    {
      "completion_length": 109.03125,
      "epoch": 1.2462927143778209,
      "grad_norm": 5.873917102813721,
      "kl": 0.140380859375,
      "learning_rate": 3.768536428110896e-07,
      "loss": 0.0056,
      "reward": 1.4805980324745178,
      "reward_std": 0.08332928642630577,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.480598047375679,
      "step": 1933
    },
    {
      "completion_length": 112.921875,
      "epoch": 1.246937459703417,
      "grad_norm": 54.74165725708008,
      "kl": 0.162841796875,
      "learning_rate": 3.765312701482914e-07,
      "loss": 0.0065,
      "reward": 1.706473171710968,
      "reward_std": 0.09516463428735733,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7064732313156128,
      "step": 1934
    },
    {
      "completion_length": 107.96875,
      "epoch": 1.2475822050290135,
      "grad_norm": 22.8502254486084,
      "kl": 0.10888671875,
      "learning_rate": 3.762088974854932e-07,
      "loss": 0.0044,
      "reward": 1.5394970774650574,
      "reward_std": 0.08196517452597618,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5394971370697021,
      "step": 1935
    },
    {
      "completion_length": 108.953125,
      "epoch": 1.24822695035461,
      "grad_norm": 13.624421119689941,
      "kl": 0.125,
      "learning_rate": 3.75886524822695e-07,
      "loss": 0.005,
      "reward": 1.4955153465270996,
      "reward_std": 0.0882682166993618,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4955153614282608,
      "step": 1936
    },
    {
      "completion_length": 98.75,
      "epoch": 1.2488716956802064,
      "grad_norm": 12.699457168579102,
      "kl": 0.114501953125,
      "learning_rate": 3.7556415215989683e-07,
      "loss": 0.0046,
      "reward": 1.7613229155540466,
      "reward_std": 0.05144652538001537,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7613229155540466,
      "step": 1937
    },
    {
      "completion_length": 113.28125,
      "epoch": 1.2495164410058026,
      "grad_norm": 233.15621948242188,
      "kl": 0.10888671875,
      "learning_rate": 3.752417794970986e-07,
      "loss": 0.0044,
      "reward": 1.5607260465621948,
      "reward_std": 0.10854167118668556,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5607260465621948,
      "step": 1938
    },
    {
      "completion_length": 102.078125,
      "epoch": 1.250161186331399,
      "grad_norm": 12.188179969787598,
      "kl": 0.115234375,
      "learning_rate": 3.7491940683430044e-07,
      "loss": 0.0046,
      "reward": 1.6384233236312866,
      "reward_std": 0.0817994475364685,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6384232938289642,
      "step": 1939
    },
    {
      "completion_length": 107.4375,
      "epoch": 1.2508059316569955,
      "grad_norm": 13.628211975097656,
      "kl": 0.123291015625,
      "learning_rate": 3.745970341715022e-07,
      "loss": 0.0049,
      "reward": 1.6886098980903625,
      "reward_std": 0.09688212350010872,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6886099278926849,
      "step": 1940
    },
    {
      "completion_length": 114.46875,
      "epoch": 1.251450676982592,
      "grad_norm": 25.029949188232422,
      "kl": 0.12646484375,
      "learning_rate": 3.7427466150870406e-07,
      "loss": 0.005,
      "reward": 1.6348047256469727,
      "reward_std": 0.11835089325904846,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6348046958446503,
      "step": 1941
    },
    {
      "completion_length": 115.5625,
      "epoch": 1.2520954223081882,
      "grad_norm": 8.084458351135254,
      "kl": 0.108642578125,
      "learning_rate": 3.7395228884590584e-07,
      "loss": 0.0043,
      "reward": 1.6494329571723938,
      "reward_std": 0.06459452584385872,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6494330167770386,
      "step": 1942
    },
    {
      "completion_length": 116.90625,
      "epoch": 1.2527401676337846,
      "grad_norm": 12.821890830993652,
      "kl": 0.10595703125,
      "learning_rate": 3.736299161831076e-07,
      "loss": 0.0042,
      "reward": 1.5637995600700378,
      "reward_std": 0.06873523443937302,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5637995451688766,
      "step": 1943
    },
    {
      "completion_length": 121.9375,
      "epoch": 1.253384912959381,
      "grad_norm": 10.103633880615234,
      "kl": 0.1142578125,
      "learning_rate": 3.7330754352030945e-07,
      "loss": 0.0046,
      "reward": 1.4810506105422974,
      "reward_std": 0.08668520301580429,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48105068504810333,
      "step": 1944
    },
    {
      "completion_length": 121.0,
      "epoch": 1.2540296582849775,
      "grad_norm": 12.35737419128418,
      "kl": 0.11767578125,
      "learning_rate": 3.7298517085751123e-07,
      "loss": 0.0047,
      "reward": 1.585673987865448,
      "reward_std": 0.12142669595777988,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5856739580631256,
      "step": 1945
    },
    {
      "completion_length": 117.4375,
      "epoch": 1.254674403610574,
      "grad_norm": 11.945507049560547,
      "kl": 0.1455078125,
      "learning_rate": 3.7266279819471306e-07,
      "loss": 0.0058,
      "reward": 1.4567408561706543,
      "reward_std": 0.1808721274137497,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.47236596047878265,
      "step": 1946
    },
    {
      "completion_length": 103.265625,
      "epoch": 1.2553191489361701,
      "grad_norm": 18.38748550415039,
      "kl": 0.1181640625,
      "learning_rate": 3.7234042553191484e-07,
      "loss": 0.0047,
      "reward": 1.580788016319275,
      "reward_std": 0.08059585839509964,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5807879716157913,
      "step": 1947
    },
    {
      "completion_length": 110.203125,
      "epoch": 1.2559638942617666,
      "grad_norm": 20.399362564086914,
      "kl": 0.116943359375,
      "learning_rate": 3.720180528691167e-07,
      "loss": 0.0047,
      "reward": 1.5159632563591003,
      "reward_std": 0.09451697021722794,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5159633159637451,
      "step": 1948
    },
    {
      "completion_length": 112.65625,
      "epoch": 1.256608639587363,
      "grad_norm": 21.889610290527344,
      "kl": 0.104736328125,
      "learning_rate": 3.7169568020631846e-07,
      "loss": 0.0042,
      "reward": 1.558445394039154,
      "reward_std": 0.11743486672639847,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5584454238414764,
      "step": 1949
    },
    {
      "completion_length": 120.78125,
      "epoch": 1.2572533849129595,
      "grad_norm": 12.497200965881348,
      "kl": 0.1220703125,
      "learning_rate": 3.713733075435203e-07,
      "loss": 0.0049,
      "reward": 1.7180161476135254,
      "reward_std": 0.07873895950615406,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7180160582065582,
      "step": 1950
    },
    {
      "completion_length": 114.359375,
      "epoch": 1.257898130238556,
      "grad_norm": 41.40201187133789,
      "kl": 0.13134765625,
      "learning_rate": 3.7105093488072207e-07,
      "loss": 0.0053,
      "reward": 1.5681995153427124,
      "reward_std": 0.06578772515058517,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5681995451450348,
      "step": 1951
    },
    {
      "completion_length": 115.03125,
      "epoch": 1.2585428755641521,
      "grad_norm": 14.451946258544922,
      "kl": 0.1181640625,
      "learning_rate": 3.707285622179239e-07,
      "loss": 0.0047,
      "reward": 1.5138095617294312,
      "reward_std": 0.17110786586999893,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.5294344574213028,
      "step": 1952
    },
    {
      "completion_length": 106.9375,
      "epoch": 1.2591876208897486,
      "grad_norm": 14.541790008544922,
      "kl": 0.126220703125,
      "learning_rate": 3.704061895551257e-07,
      "loss": 0.0051,
      "reward": 1.6685959696769714,
      "reward_std": 0.10416876897215843,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6842210292816162,
      "step": 1953
    },
    {
      "completion_length": 105.984375,
      "epoch": 1.259832366215345,
      "grad_norm": 10.058271408081055,
      "kl": 0.124755859375,
      "learning_rate": 3.700838168923275e-07,
      "loss": 0.005,
      "reward": 1.6186620593070984,
      "reward_std": 0.11036109924316406,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6186620444059372,
      "step": 1954
    },
    {
      "completion_length": 95.25,
      "epoch": 1.2604771115409412,
      "grad_norm": 9.15190315246582,
      "kl": 0.1474609375,
      "learning_rate": 3.697614442295293e-07,
      "loss": 0.0059,
      "reward": 1.6321035623550415,
      "reward_std": 0.10482674837112427,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6321035325527191,
      "step": 1955
    },
    {
      "completion_length": 100.53125,
      "epoch": 1.2611218568665377,
      "grad_norm": 10.478346824645996,
      "kl": 0.146484375,
      "learning_rate": 3.694390715667311e-07,
      "loss": 0.0059,
      "reward": 1.7911347150802612,
      "reward_std": 0.06305089220404625,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.791134774684906,
      "step": 1956
    },
    {
      "completion_length": 114.5625,
      "epoch": 1.261766602192134,
      "grad_norm": 17.663482666015625,
      "kl": 0.11083984375,
      "learning_rate": 3.691166989039329e-07,
      "loss": 0.0044,
      "reward": 1.5123714208602905,
      "reward_std": 0.10252119600772858,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5123714208602905,
      "step": 1957
    },
    {
      "completion_length": 117.234375,
      "epoch": 1.2624113475177305,
      "grad_norm": 6.887454509735107,
      "kl": 0.112548828125,
      "learning_rate": 3.687943262411347e-07,
      "loss": 0.0045,
      "reward": 1.6729698181152344,
      "reward_std": 0.07313475385308266,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6729699075222015,
      "step": 1958
    },
    {
      "completion_length": 113.265625,
      "epoch": 1.263056092843327,
      "grad_norm": 18.4481201171875,
      "kl": 0.1220703125,
      "learning_rate": 3.6847195357833653e-07,
      "loss": 0.0049,
      "reward": 1.4612666964530945,
      "reward_std": 0.07354892790317535,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4612666964530945,
      "step": 1959
    },
    {
      "completion_length": 110.921875,
      "epoch": 1.2637008381689232,
      "grad_norm": 37.00850296020508,
      "kl": 0.1123046875,
      "learning_rate": 3.681495809155383e-07,
      "loss": 0.0045,
      "reward": 1.581781268119812,
      "reward_std": 0.08708008006215096,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5817812830209732,
      "step": 1960
    },
    {
      "completion_length": 108.765625,
      "epoch": 1.2643455834945196,
      "grad_norm": 20.20766830444336,
      "kl": 0.126953125,
      "learning_rate": 3.6782720825274014e-07,
      "loss": 0.0051,
      "reward": 1.7111400365829468,
      "reward_std": 0.08487332798540592,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7111401557922363,
      "step": 1961
    },
    {
      "completion_length": 128.984375,
      "epoch": 1.264990328820116,
      "grad_norm": 14.24345588684082,
      "kl": 0.1005859375,
      "learning_rate": 3.675048355899419e-07,
      "loss": 0.004,
      "reward": 1.4533016085624695,
      "reward_std": 0.08253934234380722,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4533015191555023,
      "step": 1962
    },
    {
      "completion_length": 117.53125,
      "epoch": 1.2656350741457125,
      "grad_norm": 6.561637878417969,
      "kl": 0.107177734375,
      "learning_rate": 3.6718246292714376e-07,
      "loss": 0.0043,
      "reward": 1.3847861886024475,
      "reward_std": 0.07416202872991562,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.3847862035036087,
      "step": 1963
    },
    {
      "completion_length": 117.140625,
      "epoch": 1.266279819471309,
      "grad_norm": 13.04010009765625,
      "kl": 0.12109375,
      "learning_rate": 3.6686009026434554e-07,
      "loss": 0.0048,
      "reward": 1.555762767791748,
      "reward_std": 0.08999788016080856,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.555762767791748,
      "step": 1964
    },
    {
      "completion_length": 126.1875,
      "epoch": 1.2669245647969052,
      "grad_norm": 7.148639678955078,
      "kl": 0.1572265625,
      "learning_rate": 3.6653771760154737e-07,
      "loss": 0.0063,
      "reward": 1.5946932435035706,
      "reward_std": 0.08358455076813698,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5946933031082153,
      "step": 1965
    },
    {
      "completion_length": 117.359375,
      "epoch": 1.2675693101225016,
      "grad_norm": 16.43146324157715,
      "kl": 0.102783203125,
      "learning_rate": 3.6621534493874915e-07,
      "loss": 0.0041,
      "reward": 1.6910881996154785,
      "reward_std": 0.07531457021832466,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6910882294178009,
      "step": 1966
    },
    {
      "completion_length": 113.46875,
      "epoch": 1.268214055448098,
      "grad_norm": 12.970282554626465,
      "kl": 0.115966796875,
      "learning_rate": 3.65892972275951e-07,
      "loss": 0.0046,
      "reward": 1.5166686177253723,
      "reward_std": 0.10853827744722366,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5166686773300171,
      "step": 1967
    },
    {
      "completion_length": 102.703125,
      "epoch": 1.2688588007736943,
      "grad_norm": 10.751826286315918,
      "kl": 0.123779296875,
      "learning_rate": 3.6557059961315276e-07,
      "loss": 0.0049,
      "reward": 1.6648629307746887,
      "reward_std": 0.09130343794822693,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6648629009723663,
      "step": 1968
    },
    {
      "completion_length": 114.078125,
      "epoch": 1.2695035460992907,
      "grad_norm": 13.697860717773438,
      "kl": 0.1650390625,
      "learning_rate": 3.652482269503546e-07,
      "loss": 0.0066,
      "reward": 1.531709372997284,
      "reward_std": 0.09723073989152908,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5317093729972839,
      "step": 1969
    },
    {
      "completion_length": 131.328125,
      "epoch": 1.2701482914248872,
      "grad_norm": 13.180706024169922,
      "kl": 0.11083984375,
      "learning_rate": 3.649258542875564e-07,
      "loss": 0.0044,
      "reward": 1.6077740788459778,
      "reward_std": 0.12473616376519203,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6077740788459778,
      "step": 1970
    },
    {
      "completion_length": 119.546875,
      "epoch": 1.2707930367504836,
      "grad_norm": 12.488205909729004,
      "kl": 0.12353515625,
      "learning_rate": 3.6460348162475816e-07,
      "loss": 0.0049,
      "reward": 1.6692855954170227,
      "reward_std": 0.07351256161928177,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6692855060100555,
      "step": 1971
    },
    {
      "completion_length": 117.671875,
      "epoch": 1.27143778207608,
      "grad_norm": 7.643041610717773,
      "kl": 0.12353515625,
      "learning_rate": 3.6428110896196e-07,
      "loss": 0.0049,
      "reward": 1.496739387512207,
      "reward_std": 0.03722511604428291,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.49673932790756226,
      "step": 1972
    },
    {
      "completion_length": 113.75,
      "epoch": 1.2720825274016763,
      "grad_norm": 8.244843482971191,
      "kl": 0.1162109375,
      "learning_rate": 3.6395873629916177e-07,
      "loss": 0.0046,
      "reward": 1.5505931377410889,
      "reward_std": 0.06616784632205963,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5505931377410889,
      "step": 1973
    },
    {
      "completion_length": 114.6875,
      "epoch": 1.2727272727272727,
      "grad_norm": 9.663063049316406,
      "kl": 0.111328125,
      "learning_rate": 3.636363636363636e-07,
      "loss": 0.0045,
      "reward": 1.5648088455200195,
      "reward_std": 0.0582730658352375,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5648089051246643,
      "step": 1974
    },
    {
      "completion_length": 105.453125,
      "epoch": 1.2733720180528691,
      "grad_norm": 31.60782241821289,
      "kl": 0.109130859375,
      "learning_rate": 3.633139909735654e-07,
      "loss": 0.0044,
      "reward": 1.503617525100708,
      "reward_std": 0.07258416526019573,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.503617525100708,
      "step": 1975
    },
    {
      "completion_length": 111.046875,
      "epoch": 1.2740167633784656,
      "grad_norm": 18.72150993347168,
      "kl": 0.111572265625,
      "learning_rate": 3.629916183107672e-07,
      "loss": 0.0045,
      "reward": 1.7879358530044556,
      "reward_std": 0.10292981564998627,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7879358530044556,
      "step": 1976
    },
    {
      "completion_length": 119.375,
      "epoch": 1.274661508704062,
      "grad_norm": 16.401723861694336,
      "kl": 0.1171875,
      "learning_rate": 3.62669245647969e-07,
      "loss": 0.0047,
      "reward": 1.7952114343643188,
      "reward_std": 0.08389570191502571,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7952114343643188,
      "step": 1977
    },
    {
      "completion_length": 116.5625,
      "epoch": 1.2753062540296582,
      "grad_norm": 10.445584297180176,
      "kl": 0.1337890625,
      "learning_rate": 3.6234687298517083e-07,
      "loss": 0.0054,
      "reward": 1.70708167552948,
      "reward_std": 0.09822971373796463,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7070816159248352,
      "step": 1978
    },
    {
      "completion_length": 126.171875,
      "epoch": 1.2759509993552547,
      "grad_norm": 27.000473022460938,
      "kl": 0.11572265625,
      "learning_rate": 3.620245003223726e-07,
      "loss": 0.0046,
      "reward": 1.487796425819397,
      "reward_std": 0.10756743885576725,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.48779644072055817,
      "step": 1979
    },
    {
      "completion_length": 112.359375,
      "epoch": 1.2765957446808511,
      "grad_norm": 19.835025787353516,
      "kl": 0.125,
      "learning_rate": 3.617021276595745e-07,
      "loss": 0.005,
      "reward": 1.8274250626564026,
      "reward_std": 0.15155724436044693,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.8430500030517578,
      "step": 1980
    },
    {
      "completion_length": 113.203125,
      "epoch": 1.2772404900064473,
      "grad_norm": 14.025164604187012,
      "kl": 0.12060546875,
      "learning_rate": 3.6137975499677623e-07,
      "loss": 0.0048,
      "reward": 1.5321632623672485,
      "reward_std": 0.09312083013355732,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5321632027626038,
      "step": 1981
    },
    {
      "completion_length": 114.5625,
      "epoch": 1.2778852353320438,
      "grad_norm": 11.435771942138672,
      "kl": 0.109375,
      "learning_rate": 3.610573823339781e-07,
      "loss": 0.0044,
      "reward": 1.5915541052818298,
      "reward_std": 0.06063748896121979,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5915541350841522,
      "step": 1982
    },
    {
      "completion_length": 113.046875,
      "epoch": 1.2785299806576402,
      "grad_norm": 8.650049209594727,
      "kl": 0.12060546875,
      "learning_rate": 3.607350096711799e-07,
      "loss": 0.0048,
      "reward": 1.5912625789642334,
      "reward_std": 0.07665859535336494,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5912625789642334,
      "step": 1983
    },
    {
      "completion_length": 113.34375,
      "epoch": 1.2791747259832367,
      "grad_norm": 16.17777442932129,
      "kl": 0.108642578125,
      "learning_rate": 3.604126370083816e-07,
      "loss": 0.0043,
      "reward": 1.6600617170333862,
      "reward_std": 0.06648578122258186,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6600617170333862,
      "step": 1984
    },
    {
      "completion_length": 116.796875,
      "epoch": 1.2798194713088331,
      "grad_norm": 12.173382759094238,
      "kl": 0.0986328125,
      "learning_rate": 3.600902643455835e-07,
      "loss": 0.0039,
      "reward": 1.6574986577033997,
      "reward_std": 0.1398964375257492,
      "rewards/format_reward": 0.984375,
      "rewards/iou_timestamp_reward": 0.6731236279010773,
      "step": 1985
    },
    {
      "completion_length": 119.140625,
      "epoch": 1.2804642166344293,
      "grad_norm": 11.117454528808594,
      "kl": 0.1337890625,
      "learning_rate": 3.597678916827853e-07,
      "loss": 0.0053,
      "reward": 1.4467003345489502,
      "reward_std": 0.09076781012117863,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.4467003643512726,
      "step": 1986
    },
    {
      "completion_length": 111.703125,
      "epoch": 1.2811089619600258,
      "grad_norm": 17.02039337158203,
      "kl": 0.1435546875,
      "learning_rate": 3.594455190199871e-07,
      "loss": 0.0057,
      "reward": 1.6881715059280396,
      "reward_std": 0.09348884224891663,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6881715357303619,
      "step": 1987
    },
    {
      "completion_length": 114.0,
      "epoch": 1.2817537072856222,
      "grad_norm": 41.62820816040039,
      "kl": 0.11865234375,
      "learning_rate": 3.591231463571889e-07,
      "loss": 0.0047,
      "reward": 1.599281907081604,
      "reward_std": 0.05156847834587097,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.599281907081604,
      "step": 1988
    },
    {
      "completion_length": 115.234375,
      "epoch": 1.2823984526112184,
      "grad_norm": 17.14328384399414,
      "kl": 0.11083984375,
      "learning_rate": 3.5880077369439074e-07,
      "loss": 0.0044,
      "reward": 1.708803951740265,
      "reward_std": 0.11099771410226822,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7088040113449097,
      "step": 1989
    },
    {
      "completion_length": 119.671875,
      "epoch": 1.283043197936815,
      "grad_norm": 342.91497802734375,
      "kl": 0.177734375,
      "learning_rate": 3.584784010315925e-07,
      "loss": 0.0071,
      "reward": 1.5411792993545532,
      "reward_std": 0.13300441205501556,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5411793291568756,
      "step": 1990
    },
    {
      "completion_length": 113.25,
      "epoch": 1.2836879432624113,
      "grad_norm": 14.473013877868652,
      "kl": 0.1025390625,
      "learning_rate": 3.5815602836879435e-07,
      "loss": 0.0041,
      "reward": 1.6225409507751465,
      "reward_std": 0.07372570037841797,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6225408911705017,
      "step": 1991
    },
    {
      "completion_length": 110.390625,
      "epoch": 1.2843326885880078,
      "grad_norm": 10.521697998046875,
      "kl": 0.110107421875,
      "learning_rate": 3.5783365570599613e-07,
      "loss": 0.0044,
      "reward": 1.6926472783088684,
      "reward_std": 0.09197350218892097,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6926472783088684,
      "step": 1992
    },
    {
      "completion_length": 110.484375,
      "epoch": 1.2849774339136042,
      "grad_norm": 8.88347339630127,
      "kl": 0.122802734375,
      "learning_rate": 3.5751128304319796e-07,
      "loss": 0.0049,
      "reward": 1.773346483707428,
      "reward_std": 0.11623063683509827,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.773346483707428,
      "step": 1993
    },
    {
      "completion_length": 104.796875,
      "epoch": 1.2856221792392004,
      "grad_norm": 13.534448623657227,
      "kl": 0.109619140625,
      "learning_rate": 3.5718891038039974e-07,
      "loss": 0.0044,
      "reward": 1.5154983401298523,
      "reward_std": 0.05960991233587265,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5154983699321747,
      "step": 1994
    },
    {
      "completion_length": 118.21875,
      "epoch": 1.2862669245647969,
      "grad_norm": 11.80124282836914,
      "kl": 0.12451171875,
      "learning_rate": 3.568665377176016e-07,
      "loss": 0.005,
      "reward": 1.787231683731079,
      "reward_std": 0.11993472650647163,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.7872316539287567,
      "step": 1995
    },
    {
      "completion_length": 101.765625,
      "epoch": 1.2869116698903933,
      "grad_norm": 13.930368423461914,
      "kl": 0.125732421875,
      "learning_rate": 3.5654416505480336e-07,
      "loss": 0.005,
      "reward": 1.5652704238891602,
      "reward_std": 0.11598680540919304,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.5652704238891602,
      "step": 1996
    },
    {
      "completion_length": 106.140625,
      "epoch": 1.2875564152159897,
      "grad_norm": 8.300092697143555,
      "kl": 0.1728515625,
      "learning_rate": 3.5622179239200514e-07,
      "loss": 0.0069,
      "reward": 1.6255128383636475,
      "reward_std": 0.1058376170694828,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6255128383636475,
      "step": 1997
    },
    {
      "completion_length": 113.53125,
      "epoch": 1.2882011605415862,
      "grad_norm": 14.051785469055176,
      "kl": 0.134033203125,
      "learning_rate": 3.5589941972920697e-07,
      "loss": 0.0054,
      "reward": 1.6041387915611267,
      "reward_std": 0.08702468872070312,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6041387319564819,
      "step": 1998
    },
    {
      "completion_length": 111.484375,
      "epoch": 1.2888459058671824,
      "grad_norm": 14.08708667755127,
      "kl": 0.14501953125,
      "learning_rate": 3.5557704706640875e-07,
      "loss": 0.0058,
      "reward": 1.6865518689155579,
      "reward_std": 0.07037417963147163,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6865518391132355,
      "step": 1999
    },
    {
      "completion_length": 116.109375,
      "epoch": 1.2894906511927788,
      "grad_norm": 13.85488510131836,
      "kl": 0.109130859375,
      "learning_rate": 3.552546744036106e-07,
      "loss": 0.0044,
      "reward": 1.6934226155281067,
      "reward_std": 0.08548137731850147,
      "rewards/format_reward": 1.0,
      "rewards/iou_timestamp_reward": 0.6934226155281067,
      "step": 2000
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 3102,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}