{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 161,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "advantages": 0.0,
      "completion_length": 77.421875,
      "epoch": 0.006211180124223602,
      "grad_norm": 4.007043838500977,
      "kl": 0.0,
      "learning_rate": 9.937888198757763e-07,
      "loss": -0.0,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 1
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 78.921875,
      "epoch": 0.012422360248447204,
      "grad_norm": 7.15122652053833,
      "kl": 0.00041961669921875,
      "learning_rate": 9.875776397515528e-07,
      "loss": 0.0,
      "reward": 1.53125,
      "reward_mean": 1.53125,
      "reward_std": 0.2845909595489502,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 2
    },
    {
      "advantages": 0.0,
      "completion_length": 78.671875,
      "epoch": 0.018633540372670808,
      "grad_norm": 2.9171459674835205,
      "kl": 0.000392913818359375,
      "learning_rate": 9.813664596273291e-07,
      "loss": 0.0,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 3
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 77.78125,
      "epoch": 0.024844720496894408,
      "grad_norm": 5.474589824676514,
      "kl": 0.000759124755859375,
      "learning_rate": 9.751552795031055e-07,
      "loss": 0.0001,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.213067427277565,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 4
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 90.8125,
      "epoch": 0.031055900621118012,
      "grad_norm": 5.2480363845825195,
      "kl": 0.001495361328125,
      "learning_rate": 9.68944099378882e-07,
      "loss": 0.0001,
      "reward": 1.53125,
      "reward_mean": 1.53125,
      "reward_std": 0.17570313811302185,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 5
    },
    {
      "advantages": 6.51925802230835e-09,
      "completion_length": 81.203125,
      "epoch": 0.037267080745341616,
      "grad_norm": 8.329508781433105,
      "kl": 0.006256103515625,
      "learning_rate": 9.627329192546583e-07,
      "loss": 0.0006,
      "reward": 1.375,
      "reward_mean": 1.375,
      "reward_std": 0.26409146189689636,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 6
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 84.265625,
      "epoch": 0.043478260869565216,
      "grad_norm": 9.400680541992188,
      "kl": 0.0106201171875,
      "learning_rate": 9.565217391304349e-07,
      "loss": 0.0011,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.2404065877199173,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 7
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 84.109375,
      "epoch": 0.049689440993788817,
      "grad_norm": 6.011223316192627,
      "kl": 0.0057373046875,
      "learning_rate": 9.503105590062112e-07,
      "loss": 0.0006,
      "reward": 1.4375,
      "reward_mean": 1.4375,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 8
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 87.046875,
      "epoch": 0.055900621118012424,
      "grad_norm": 4.103212356567383,
      "kl": 0.00244140625,
      "learning_rate": 9.440993788819875e-07,
      "loss": 0.0002,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 9
    },
    {
      "advantages": -7.450580596923828e-09,
      "completion_length": 76.15625,
      "epoch": 0.062111801242236024,
      "grad_norm": 7.4132466316223145,
      "kl": 0.01287841796875,
      "learning_rate": 9.37888198757764e-07,
      "loss": 0.0013,
      "reward": 1.4375,
      "reward_mean": 1.4375,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 10
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 76.75,
      "epoch": 0.06832298136645963,
      "grad_norm": 4.6751017570495605,
      "kl": 0.0130615234375,
      "learning_rate": 9.316770186335403e-07,
      "loss": 0.0013,
      "reward": 1.28125,
      "reward_mean": 1.28125,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 11
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 76.390625,
      "epoch": 0.07453416149068323,
      "grad_norm": 7.682182788848877,
      "kl": 0.018310546875,
      "learning_rate": 9.254658385093167e-07,
      "loss": 0.0018,
      "reward": 1.734375,
      "reward_mean": 1.734375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 12
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 85.234375,
      "epoch": 0.08074534161490683,
      "grad_norm": 4.814305305480957,
      "kl": 0.00396728515625,
      "learning_rate": 9.19254658385093e-07,
      "loss": 0.0004,
      "reward": 1.8125,
      "reward_mean": 1.8125,
      "reward_std": 0.22461533546447754,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 13
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 86.3125,
      "epoch": 0.08695652173913043,
      "grad_norm": 189.3062744140625,
      "kl": 0.033203125,
      "learning_rate": 9.130434782608695e-07,
      "loss": 0.0033,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "advantages": 2.7939677238464355e-09,
      "completion_length": 69.8125,
      "epoch": 0.09316770186335403,
      "grad_norm": 3.5520412921905518,
      "kl": 0.01129150390625,
      "learning_rate": 9.06832298136646e-07,
      "loss": 0.0011,
      "reward": 1.78125,
      "reward_mean": 1.78125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 76.296875,
      "epoch": 0.09937888198757763,
      "grad_norm": 3.526542901992798,
      "kl": 0.00909423828125,
      "learning_rate": 9.006211180124223e-07,
      "loss": 0.0009,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 16
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 86.3125,
      "epoch": 0.10559006211180125,
      "grad_norm": 5.98048210144043,
      "kl": 0.0057373046875,
      "learning_rate": 8.944099378881988e-07,
      "loss": 0.0006,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "advantages": 0.0,
      "completion_length": 82.921875,
      "epoch": 0.11180124223602485,
      "grad_norm": 2.0705599784851074,
      "kl": 0.00592041015625,
      "learning_rate": 8.881987577639751e-07,
      "loss": 0.0006,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.984375,
      "step": 18
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 81.40625,
      "epoch": 0.11801242236024845,
      "grad_norm": 9.266715049743652,
      "kl": 0.0079345703125,
      "learning_rate": 8.819875776397515e-07,
      "loss": 0.0008,
      "reward": 1.546875,
      "reward_mean": 1.546875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.984375,
      "step": 19
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.203125,
      "epoch": 0.12422360248447205,
      "grad_norm": 10.367863655090332,
      "kl": 0.0072021484375,
      "learning_rate": 8.757763975155279e-07,
      "loss": 0.0007,
      "reward": 1.40625,
      "reward_mean": 1.40625,
      "reward_std": 0.2404065728187561,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 75.734375,
      "epoch": 0.13043478260869565,
      "grad_norm": 2.6553478240966797,
      "kl": 0.00592041015625,
      "learning_rate": 8.695652173913043e-07,
      "loss": 0.0006,
      "reward": 1.578125,
      "reward_mean": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 85.8125,
      "epoch": 0.13664596273291926,
      "grad_norm": 3.458266496658325,
      "kl": 0.00604248046875,
      "learning_rate": 8.633540372670807e-07,
      "loss": 0.0006,
      "reward": 1.515625,
      "reward_mean": 1.515625,
      "reward_std": 0.15981829166412354,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.984375,
      "step": 22
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 79.328125,
      "epoch": 0.14285714285714285,
      "grad_norm": 3.2002384662628174,
      "kl": 0.00543212890625,
      "learning_rate": 8.57142857142857e-07,
      "loss": 0.0005,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 23
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 75.375,
      "epoch": 0.14906832298136646,
      "grad_norm": 5.946903705596924,
      "kl": 0.0087890625,
      "learning_rate": 8.509316770186336e-07,
      "loss": 0.0009,
      "reward": 1.484375,
      "reward_mean": 1.484375,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.984375,
      "step": 24
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 68.8125,
      "epoch": 0.15527950310559005,
      "grad_norm": 4.977855682373047,
      "kl": 0.008544921875,
      "learning_rate": 8.447204968944099e-07,
      "loss": 0.0009,
      "reward": 1.734375,
      "reward_mean": 1.734375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "advantages": -8.381903171539307e-09,
      "completion_length": 83.59375,
      "epoch": 0.16149068322981366,
      "grad_norm": 4.409206390380859,
      "kl": 0.01239013671875,
      "learning_rate": 8.385093167701863e-07,
      "loss": 0.0012,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 78.109375,
      "epoch": 0.16770186335403728,
      "grad_norm": 3.1185989379882812,
      "kl": 0.006072998046875,
      "learning_rate": 8.322981366459628e-07,
      "loss": 0.0006,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 27
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 72.0625,
      "epoch": 0.17391304347826086,
      "grad_norm": 4.9565935134887695,
      "kl": 0.010009765625,
      "learning_rate": 8.260869565217391e-07,
      "loss": 0.001,
      "reward": 1.34375,
      "reward_mean": 1.34375,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 28
    },
    {
      "advantages": -4.6566128730773926e-09,
      "completion_length": 84.953125,
      "epoch": 0.18012422360248448,
      "grad_norm": 4.35609769821167,
      "kl": 0.011962890625,
      "learning_rate": 8.198757763975155e-07,
      "loss": 0.0012,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.25513991713523865,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.984375,
      "step": 29
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 85.34375,
      "epoch": 0.18633540372670807,
      "grad_norm": 5.767938137054443,
      "kl": 0.009033203125,
      "learning_rate": 8.136645962732918e-07,
      "loss": 0.0009,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 30
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 83.5625,
      "epoch": 0.19254658385093168,
      "grad_norm": 49.12059783935547,
      "kl": 0.0091552734375,
      "learning_rate": 8.074534161490683e-07,
      "loss": 0.0009,
      "reward": 1.578125,
      "reward_mean": 1.578125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 31
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 77.734375,
      "epoch": 0.19875776397515527,
      "grad_norm": 1.4828208684921265,
      "kl": 0.00970458984375,
      "learning_rate": 8.012422360248446e-07,
      "loss": 0.001,
      "reward": 1.421875,
      "reward_mean": 1.421875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 32
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 78.453125,
      "epoch": 0.20496894409937888,
      "grad_norm": 7.876468658447266,
      "kl": 0.020263671875,
      "learning_rate": 7.95031055900621e-07,
      "loss": 0.002,
      "reward": 1.734375,
      "reward_mean": 1.734375,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 33
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.5,
      "epoch": 0.2111801242236025,
      "grad_norm": 3.8213541507720947,
      "kl": 0.01361083984375,
      "learning_rate": 7.888198757763976e-07,
      "loss": 0.0014,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "advantages": -6.51925802230835e-09,
      "completion_length": 89.5625,
      "epoch": 0.21739130434782608,
      "grad_norm": 3.453101634979248,
      "kl": 0.01470947265625,
      "learning_rate": 7.826086956521739e-07,
      "loss": 0.0015,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "advantages": 0.0,
      "completion_length": 90.28125,
      "epoch": 0.2236024844720497,
      "grad_norm": 2.642101526260376,
      "kl": 0.0107421875,
      "learning_rate": 7.763975155279503e-07,
      "loss": 0.0011,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 36
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.3125,
      "epoch": 0.22981366459627328,
      "grad_norm": 3.5424673557281494,
      "kl": 0.01239013671875,
      "learning_rate": 7.701863354037266e-07,
      "loss": 0.0012,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "advantages": 0.0,
      "completion_length": 84.140625,
      "epoch": 0.2360248447204969,
      "grad_norm": 0.38800248503685,
      "kl": 0.01275634765625,
      "learning_rate": 7.639751552795031e-07,
      "loss": 0.0013,
      "reward": 1.375,
      "reward_mean": 1.375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.984375,
      "step": 38
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 89.109375,
      "epoch": 0.2422360248447205,
      "grad_norm": 2.645474433898926,
      "kl": 0.01397705078125,
      "learning_rate": 7.577639751552795e-07,
      "loss": 0.0014,
      "reward": 1.515625,
      "reward_mean": 1.515625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 72.296875,
      "epoch": 0.2484472049689441,
      "grad_norm": 8.574762344360352,
      "kl": 0.0159912109375,
      "learning_rate": 7.515527950310558e-07,
      "loss": 0.0016,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 86.046875,
      "epoch": 0.2546583850931677,
      "grad_norm": 36.26329040527344,
      "kl": 0.0147705078125,
      "learning_rate": 7.453416149068323e-07,
      "loss": 0.0015,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 41
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 77.0625,
      "epoch": 0.2608695652173913,
      "grad_norm": 10.992830276489258,
      "kl": 0.0113525390625,
      "learning_rate": 7.391304347826086e-07,
      "loss": 0.0011,
      "reward": 1.703125,
      "reward_mean": 1.703125,
      "reward_std": 0.24464011192321777,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 42
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 86.53125,
      "epoch": 0.2670807453416149,
      "grad_norm": 6.251725673675537,
      "kl": 0.009033203125,
      "learning_rate": 7.329192546583851e-07,
      "loss": 0.0009,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "advantages": -4.6566128730773926e-09,
      "completion_length": 86.4375,
      "epoch": 0.2732919254658385,
      "grad_norm": 3.8048486709594727,
      "kl": 0.01385498046875,
      "learning_rate": 7.267080745341615e-07,
      "loss": 0.0014,
      "reward": 1.765625,
      "reward_mean": 1.765625,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 44
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 84.21875,
      "epoch": 0.2795031055900621,
      "grad_norm": 2.5062499046325684,
      "kl": 0.00811767578125,
      "learning_rate": 7.204968944099379e-07,
      "loss": 0.0008,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.984375,
      "step": 45
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 77.9375,
      "epoch": 0.2857142857142857,
      "grad_norm": 3.8415560722351074,
      "kl": 0.01165771484375,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.0012,
      "reward": 1.53125,
      "reward_mean": 1.53125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 46
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 82.6875,
      "epoch": 0.2919254658385093,
      "grad_norm": 2.903069496154785,
      "kl": 0.012451171875,
      "learning_rate": 7.080745341614906e-07,
      "loss": 0.0012,
      "reward": 1.578125,
      "reward_mean": 1.578125,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.984375,
      "step": 47
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 75.578125,
      "epoch": 0.2981366459627329,
      "grad_norm": 11.884781837463379,
      "kl": 0.0125732421875,
      "learning_rate": 7.018633540372671e-07,
      "loss": 0.0013,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.17570312321186066,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 73.34375,
      "epoch": 0.30434782608695654,
      "grad_norm": 2.234876871109009,
      "kl": 0.0084228515625,
      "learning_rate": 6.956521739130434e-07,
      "loss": 0.0008,
      "reward": 1.484375,
      "reward_mean": 1.484375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 81.8125,
      "epoch": 0.3105590062111801,
      "grad_norm": 4.401739597320557,
      "kl": 0.007232666015625,
      "learning_rate": 6.894409937888198e-07,
      "loss": 0.0007,
      "reward": 1.765625,
      "reward_mean": 1.765625,
      "reward_std": 0.17782479524612427,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "advantages": 0.0,
      "completion_length": 84.015625,
      "epoch": 0.3167701863354037,
      "grad_norm": 0.28293830156326294,
      "kl": 0.0062255859375,
      "learning_rate": 6.832298136645962e-07,
      "loss": 0.0006,
      "reward": 2.0,
      "reward_mean": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "advantages": 0.0,
      "completion_length": 79.125,
      "epoch": 0.32298136645962733,
      "grad_norm": 2.2039246559143066,
      "kl": 0.0106201171875,
      "learning_rate": 6.770186335403726e-07,
      "loss": 0.0011,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 52
    },
    {
      "advantages": 9.313225746154785e-10,
      "completion_length": 76.0625,
      "epoch": 0.32919254658385094,
      "grad_norm": 4.176709175109863,
      "kl": 0.01123046875,
      "learning_rate": 6.708074534161491e-07,
      "loss": 0.0011,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 81.125,
      "epoch": 0.33540372670807456,
      "grad_norm": 30.12848663330078,
      "kl": 0.099609375,
      "learning_rate": 6.645962732919254e-07,
      "loss": 0.01,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 80.609375,
      "epoch": 0.3416149068322981,
      "grad_norm": 12.808406829833984,
      "kl": 0.01416015625,
      "learning_rate": 6.583850931677019e-07,
      "loss": 0.0014,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.2238783985376358,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 55
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 76.15625,
      "epoch": 0.34782608695652173,
      "grad_norm": 5.750046253204346,
      "kl": 0.01019287109375,
      "learning_rate": 6.521739130434782e-07,
      "loss": 0.001,
      "reward": 1.5,
      "reward_mean": 1.5,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.984375,
      "step": 56
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 76.765625,
      "epoch": 0.35403726708074534,
      "grad_norm": 4.7853102684021,
      "kl": 0.010986328125,
      "learning_rate": 6.459627329192546e-07,
      "loss": 0.0011,
      "reward": 1.328125,
      "reward_mean": 1.328125,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 0.984375,
      "step": 57
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 88.796875,
      "epoch": 0.36024844720496896,
      "grad_norm": 1.7344197034835815,
      "kl": 0.00982666015625,
      "learning_rate": 6.39751552795031e-07,
      "loss": 0.001,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 58
    },
    {
      "advantages": -9.313225746154785e-09,
      "completion_length": 84.28125,
      "epoch": 0.36645962732919257,
      "grad_norm": 3.1260499954223633,
      "kl": 0.01416015625,
      "learning_rate": 6.335403726708074e-07,
      "loss": 0.0014,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.1828794628381729,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.984375,
      "step": 59
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 77.21875,
      "epoch": 0.37267080745341613,
      "grad_norm": 1.7963190078735352,
      "kl": 0.0089111328125,
      "learning_rate": 6.273291925465838e-07,
      "loss": 0.0009,
      "reward": 1.84375,
      "reward_mean": 1.84375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "advantages": 0.0,
      "completion_length": 82.25,
      "epoch": 0.37888198757763975,
      "grad_norm": 2.5049538612365723,
      "kl": 0.00787353515625,
      "learning_rate": 6.211180124223601e-07,
      "loss": 0.0008,
      "reward": 1.625,
      "reward_mean": 1.625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "advantages": -1.210719347000122e-08,
      "completion_length": 80.375,
      "epoch": 0.38509316770186336,
      "grad_norm": 5.739541530609131,
      "kl": 0.01312255859375,
      "learning_rate": 6.149068322981367e-07,
      "loss": 0.0013,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 84.296875,
      "epoch": 0.391304347826087,
      "grad_norm": 4.335031032562256,
      "kl": 0.01116943359375,
      "learning_rate": 6.08695652173913e-07,
      "loss": 0.0011,
      "reward": 1.90625,
      "reward_mean": 1.90625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 63
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 86.828125,
      "epoch": 0.39751552795031053,
      "grad_norm": 4.443232536315918,
      "kl": 0.01336669921875,
      "learning_rate": 6.024844720496894e-07,
      "loss": 0.0013,
      "reward": 1.703125,
      "reward_mean": 1.703125,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "advantages": 5.587935447692871e-09,
      "completion_length": 75.71875,
      "epoch": 0.40372670807453415,
      "grad_norm": 7.092515468597412,
      "kl": 0.01251220703125,
      "learning_rate": 5.962732919254659e-07,
      "loss": 0.0013,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.23827511072158813,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 82.140625,
      "epoch": 0.40993788819875776,
      "grad_norm": 4.468729496002197,
      "kl": 0.0211181640625,
      "learning_rate": 5.900621118012422e-07,
      "loss": 0.0021,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 66
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 74.890625,
      "epoch": 0.4161490683229814,
      "grad_norm": 9.289567947387695,
      "kl": 0.01611328125,
      "learning_rate": 5.838509316770186e-07,
      "loss": 0.0016,
      "reward": 1.421875,
      "reward_mean": 1.421875,
      "reward_std": 0.1983242630958557,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 67
    },
    {
      "advantages": 0.0,
      "completion_length": 77.625,
      "epoch": 0.422360248447205,
      "grad_norm": 0.4326918125152588,
      "kl": 0.0140380859375,
      "learning_rate": 5.77639751552795e-07,
      "loss": 0.0014,
      "reward": 1.875,
      "reward_mean": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 81.71875,
      "epoch": 0.42857142857142855,
      "grad_norm": 5.539842128753662,
      "kl": 0.04296875,
      "learning_rate": 5.714285714285714e-07,
      "loss": 0.0043,
      "reward": 1.4375,
      "reward_mean": 1.4375,
      "reward_std": 0.34352827072143555,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.984375,
      "step": 69
    },
    {
      "advantages": 0.0,
      "completion_length": 90.703125,
      "epoch": 0.43478260869565216,
      "grad_norm": 0.46686819195747375,
      "kl": 0.0074462890625,
      "learning_rate": 5.652173913043477e-07,
      "loss": 0.0007,
      "reward": 1.875,
      "reward_mean": 1.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 83.578125,
      "epoch": 0.4409937888198758,
      "grad_norm": 8.54028606414795,
      "kl": 0.00897216796875,
      "learning_rate": 5.590062111801241e-07,
      "loss": 0.0009,
      "reward": 1.765625,
      "reward_mean": 1.765625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 84.25,
      "epoch": 0.4472049689440994,
      "grad_norm": 12.895256996154785,
      "kl": 0.00579833984375,
      "learning_rate": 5.527950310559007e-07,
      "loss": 0.0006,
      "reward": 1.453125,
      "reward_mean": 1.453125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 72
    },
    {
      "advantages": -9.313225746154785e-09,
      "completion_length": 77.15625,
      "epoch": 0.453416149068323,
      "grad_norm": 5.548634052276611,
      "kl": 0.0123291015625,
      "learning_rate": 5.46583850931677e-07,
      "loss": 0.0012,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.31983357667922974,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "advantages": -1.0244548320770264e-08,
      "completion_length": 84.75,
      "epoch": 0.45962732919254656,
      "grad_norm": 3.4154112339019775,
      "kl": 0.018798828125,
      "learning_rate": 5.403726708074534e-07,
      "loss": 0.0019,
      "reward": 1.78125,
      "reward_mean": 1.78125,
      "reward_std": 0.19727616012096405,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 74
    },
    {
      "advantages": -4.6566128730773926e-09,
      "completion_length": 83.015625,
      "epoch": 0.4658385093167702,
      "grad_norm": 3.4328691959381104,
      "kl": 0.01275634765625,
      "learning_rate": 5.341614906832298e-07,
      "loss": 0.0013,
      "reward": 1.53125,
      "reward_mean": 1.53125,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 78.609375,
      "epoch": 0.4720496894409938,
      "grad_norm": 3.627190113067627,
      "kl": 0.0142822265625,
      "learning_rate": 5.279503105590062e-07,
      "loss": 0.0014,
      "reward": 1.9375,
      "reward_mean": 1.9375,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 76
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.46875,
      "epoch": 0.4782608695652174,
      "grad_norm": 10.168981552124023,
      "kl": 0.01251220703125,
      "learning_rate": 5.217391304347825e-07,
      "loss": 0.0013,
      "reward": 1.515625,
      "reward_mean": 1.515625,
      "reward_std": 0.2109457552433014,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "advantages": 0.0,
      "completion_length": 83.421875,
      "epoch": 0.484472049689441,
      "grad_norm": 20.923242568969727,
      "kl": 0.0113525390625,
      "learning_rate": 5.15527950310559e-07,
      "loss": 0.0011,
      "reward": 1.828125,
      "reward_mean": 1.828125,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 78
    },
    {
      "advantages": 2.7939677238464355e-09,
      "completion_length": 75.109375,
      "epoch": 0.4906832298136646,
      "grad_norm": 3.643770933151245,
      "kl": 0.00811767578125,
      "learning_rate": 5.093167701863354e-07,
      "loss": 0.0008,
      "reward": 1.78125,
      "reward_mean": 1.78125,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 81.65625,
      "epoch": 0.4968944099378882,
      "grad_norm": 4.883938312530518,
      "kl": 0.015625,
      "learning_rate": 5.031055900621117e-07,
      "loss": 0.0016,
      "reward": 1.25,
      "reward_mean": 1.25,
      "reward_std": 0.2130674123764038,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 80
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 82.515625,
      "epoch": 0.5031055900621118,
      "grad_norm": 1.3860398530960083,
      "kl": 0.00799560546875,
      "learning_rate": 4.968944099378881e-07,
      "loss": 0.0008,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 81
    },
    {
      "advantages": 9.313225746154785e-10,
      "completion_length": 77.6875,
      "epoch": 0.5093167701863354,
      "grad_norm": 3.7328872680664062,
      "kl": 0.0181884765625,
      "learning_rate": 4.906832298136646e-07,
      "loss": 0.0018,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 82
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 76.4375,
      "epoch": 0.515527950310559,
      "grad_norm": 3.6228644847869873,
      "kl": 0.01446533203125,
      "learning_rate": 4.84472049689441e-07,
      "loss": 0.0014,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 83
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 79.140625,
      "epoch": 0.5217391304347826,
      "grad_norm": 5.579171180725098,
      "kl": 0.0159912109375,
      "learning_rate": 4.782608695652174e-07,
      "loss": 0.0016,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.23356688022613525,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 84
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 80.0,
      "epoch": 0.5279503105590062,
      "grad_norm": 9.611387252807617,
      "kl": 0.01080322265625,
      "learning_rate": 4.7204968944099376e-07,
      "loss": 0.0011,
      "reward": 1.828125,
      "reward_mean": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 85.5,
      "epoch": 0.5341614906832298,
      "grad_norm": 4.1448540687561035,
      "kl": 0.01007080078125,
      "learning_rate": 4.6583850931677014e-07,
      "loss": 0.001,
      "reward": 1.859375,
      "reward_mean": 1.859375,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 75.5,
      "epoch": 0.5403726708074534,
      "grad_norm": 5.654483795166016,
      "kl": 0.01123046875,
      "learning_rate": 4.596273291925465e-07,
      "loss": 0.0011,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 87
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 72.671875,
      "epoch": 0.546583850931677,
      "grad_norm": 2.2370052337646484,
      "kl": 0.0137939453125,
      "learning_rate": 4.53416149068323e-07,
      "loss": 0.0014,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "advantages": 0.0,
      "completion_length": 81.84375,
      "epoch": 0.5527950310559007,
      "grad_norm": 1.389394760131836,
      "kl": 0.00836181640625,
      "learning_rate": 4.472049689440994e-07,
      "loss": 0.0008,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.984375,
      "step": 89
    },
    {
      "advantages": 0.0,
      "completion_length": 74.59375,
      "epoch": 0.5590062111801242,
      "grad_norm": 2.353760242462158,
      "kl": 0.00811767578125,
      "learning_rate": 4.4099378881987576e-07,
      "loss": 0.0008,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "advantages": 0.0,
      "completion_length": 85.28125,
      "epoch": 0.5652173913043478,
      "grad_norm": 1.5767848491668701,
      "kl": 0.009765625,
      "learning_rate": 4.3478260869565214e-07,
      "loss": 0.001,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "advantages": -8.381903171539307e-09,
      "completion_length": 81.859375,
      "epoch": 0.5714285714285714,
      "grad_norm": 3.835320234298706,
      "kl": 0.0181884765625,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0018,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.984375,
      "step": 92
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 83.671875,
      "epoch": 0.577639751552795,
      "grad_norm": 9.30271053314209,
      "kl": 0.017822265625,
      "learning_rate": 4.2236024844720495e-07,
      "loss": 0.0018,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "advantages": -4.6566128730773926e-09,
      "completion_length": 77.53125,
      "epoch": 0.5838509316770186,
      "grad_norm": 6.170975685119629,
      "kl": 0.009521484375,
      "learning_rate": 4.161490683229814e-07,
      "loss": 0.001,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 85.640625,
      "epoch": 0.5900621118012422,
      "grad_norm": 4.217593669891357,
      "kl": 0.01409912109375,
      "learning_rate": 4.0993788819875776e-07,
      "loss": 0.0014,
      "reward": 1.734375,
      "reward_mean": 1.734375,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.984375,
      "step": 95
    },
    {
      "advantages": 0.0,
      "completion_length": 77.296875,
      "epoch": 0.5962732919254659,
      "grad_norm": 6.138365268707275,
      "kl": 0.0106201171875,
      "learning_rate": 4.0372670807453413e-07,
      "loss": 0.0011,
      "reward": 1.375,
      "reward_mean": 1.375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.984375,
      "step": 96
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 76.484375,
      "epoch": 0.6024844720496895,
      "grad_norm": 1.2896429300308228,
      "kl": 0.00970458984375,
      "learning_rate": 3.975155279503105e-07,
      "loss": 0.001,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 97
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 81.71875,
      "epoch": 0.6086956521739131,
      "grad_norm": 6.941093444824219,
      "kl": 0.01165771484375,
      "learning_rate": 3.9130434782608694e-07,
      "loss": 0.0012,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 81.390625,
      "epoch": 0.6149068322981367,
      "grad_norm": 3.163457155227661,
      "kl": 0.00787353515625,
      "learning_rate": 3.850931677018633e-07,
      "loss": 0.0008,
      "reward": 1.96875,
      "reward_mean": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 99
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 79.1875,
      "epoch": 0.6211180124223602,
      "grad_norm": 4.2669830322265625,
      "kl": 0.0108642578125,
      "learning_rate": 3.7888198757763975e-07,
      "loss": 0.0011,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 77.578125,
      "epoch": 0.6273291925465838,
      "grad_norm": 6.153615474700928,
      "kl": 0.0101318359375,
      "learning_rate": 3.7267080745341613e-07,
      "loss": 0.001,
      "reward": 1.359375,
      "reward_mean": 1.359375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 101
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 81.53125,
      "epoch": 0.6335403726708074,
      "grad_norm": 4.077609539031982,
      "kl": 0.0181884765625,
      "learning_rate": 3.6645962732919256e-07,
      "loss": 0.0018,
      "reward": 1.84375,
      "reward_mean": 1.84375,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.375,
      "epoch": 0.639751552795031,
      "grad_norm": 3.084027051925659,
      "kl": 0.01007080078125,
      "learning_rate": 3.6024844720496894e-07,
      "loss": 0.001,
      "reward": 1.53125,
      "reward_mean": 1.53125,
      "reward_std": 0.1462521106004715,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 83.953125,
      "epoch": 0.6459627329192547,
      "grad_norm": 2.0512335300445557,
      "kl": 0.007476806640625,
      "learning_rate": 3.540372670807453e-07,
      "loss": 0.0007,
      "reward": 1.453125,
      "reward_mean": 1.453125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "advantages": 0.0,
      "completion_length": 82.5625,
      "epoch": 0.6521739130434783,
      "grad_norm": 0.5302374362945557,
      "kl": 0.00982666015625,
      "learning_rate": 3.478260869565217e-07,
      "loss": 0.001,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 105
    },
    {
      "advantages": -4.6566128730773926e-09,
      "completion_length": 75.953125,
      "epoch": 0.6583850931677019,
      "grad_norm": 6.2678751945495605,
      "kl": 0.0111083984375,
      "learning_rate": 3.416149068322981e-07,
      "loss": 0.0011,
      "reward": 1.890625,
      "reward_mean": 1.890625,
      "reward_std": 0.1315089464187622,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 78.328125,
      "epoch": 0.6645962732919255,
      "grad_norm": 1.7859537601470947,
      "kl": 0.00946044921875,
      "learning_rate": 3.3540372670807456e-07,
      "loss": 0.0009,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 107
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 91.8125,
      "epoch": 0.6708074534161491,
      "grad_norm": 2.7167623043060303,
      "kl": 0.0081787109375,
      "learning_rate": 3.2919254658385094e-07,
      "loss": 0.0008,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.984375,
      "step": 108
    },
    {
      "advantages": 0.0,
      "completion_length": 76.984375,
      "epoch": 0.6770186335403726,
      "grad_norm": 5.3628058433532715,
      "kl": 0.009033203125,
      "learning_rate": 3.229813664596273e-07,
      "loss": 0.0009,
      "reward": 1.515625,
      "reward_mean": 1.515625,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 73.5,
      "epoch": 0.6832298136645962,
      "grad_norm": 3.2727582454681396,
      "kl": 0.0108642578125,
      "learning_rate": 3.167701863354037e-07,
      "loss": 0.0011,
      "reward": 1.609375,
      "reward_mean": 1.609375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "advantages": 0.0,
      "completion_length": 75.75,
      "epoch": 0.6894409937888198,
      "grad_norm": 11.552366256713867,
      "kl": 0.0145263671875,
      "learning_rate": 3.105590062111801e-07,
      "loss": 0.0015,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 111
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 85.734375,
      "epoch": 0.6956521739130435,
      "grad_norm": 6.025736331939697,
      "kl": 0.01556396484375,
      "learning_rate": 3.043478260869565e-07,
      "loss": 0.0016,
      "reward": 1.59375,
      "reward_mean": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 112
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 82.734375,
      "epoch": 0.7018633540372671,
      "grad_norm": 15.336418151855469,
      "kl": 0.057373046875,
      "learning_rate": 2.9813664596273294e-07,
      "loss": 0.0057,
      "reward": 1.84375,
      "reward_mean": 1.84375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 78.953125,
      "epoch": 0.7080745341614907,
      "grad_norm": 65.76184844970703,
      "kl": 0.01385498046875,
      "learning_rate": 2.919254658385093e-07,
      "loss": 0.0014,
      "reward": 1.90625,
      "reward_mean": 1.90625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 114
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 79.765625,
      "epoch": 0.7142857142857143,
      "grad_norm": 3.660456657409668,
      "kl": 0.0194091796875,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0019,
      "reward": 1.59375,
      "reward_mean": 1.59375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 76.34375,
      "epoch": 0.7204968944099379,
      "grad_norm": 4.989613056182861,
      "kl": 0.00787353515625,
      "learning_rate": 2.7950310559006207e-07,
      "loss": 0.0008,
      "reward": 1.828125,
      "reward_mean": 1.828125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 0.984375,
      "step": 116
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 77.859375,
      "epoch": 0.7267080745341615,
      "grad_norm": 2.4932050704956055,
      "kl": 0.0081787109375,
      "learning_rate": 2.732919254658385e-07,
      "loss": 0.0008,
      "reward": 1.859375,
      "reward_mean": 1.859375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 117
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 84.5,
      "epoch": 0.7329192546583851,
      "grad_norm": 5.0420732498168945,
      "kl": 0.01226806640625,
      "learning_rate": 2.670807453416149e-07,
      "loss": 0.0012,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.23144522309303284,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 118
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 79.703125,
      "epoch": 0.7391304347826086,
      "grad_norm": 3.599855899810791,
      "kl": 0.0086669921875,
      "learning_rate": 2.6086956521739126e-07,
      "loss": 0.0009,
      "reward": 1.484375,
      "reward_mean": 1.484375,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 79.1875,
      "epoch": 0.7453416149068323,
      "grad_norm": 3.320706605911255,
      "kl": 0.0133056640625,
      "learning_rate": 2.546583850931677e-07,
      "loss": 0.0013,
      "reward": 1.828125,
      "reward_mean": 1.828125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "advantages": -8.381903171539307e-09,
      "completion_length": 95.203125,
      "epoch": 0.7515527950310559,
      "grad_norm": 2.8366851806640625,
      "kl": 0.0064697265625,
      "learning_rate": 2.4844720496894407e-07,
      "loss": 0.0006,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.1530819833278656,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 78.765625,
      "epoch": 0.7577639751552795,
      "grad_norm": 3.376732587814331,
      "kl": 0.00860595703125,
      "learning_rate": 2.422360248447205e-07,
      "loss": 0.0009,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 79.453125,
      "epoch": 0.7639751552795031,
      "grad_norm": 3.5682129859924316,
      "kl": 0.018798828125,
      "learning_rate": 2.3602484472049688e-07,
      "loss": 0.0019,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 123
    },
    {
      "advantages": 4.6566128730773926e-09,
      "completion_length": 74.96875,
      "epoch": 0.7701863354037267,
      "grad_norm": 2.6698434352874756,
      "kl": 0.006256103515625,
      "learning_rate": 2.2981366459627326e-07,
      "loss": 0.0006,
      "reward": 1.546875,
      "reward_mean": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "advantages": 0.0,
      "completion_length": 78.421875,
      "epoch": 0.7763975155279503,
      "grad_norm": 3.1063811779022217,
      "kl": 0.01214599609375,
      "learning_rate": 2.236024844720497e-07,
      "loss": 0.0012,
      "reward": 1.765625,
      "reward_mean": 1.765625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 86.703125,
      "epoch": 0.782608695652174,
      "grad_norm": 2.7392446994781494,
      "kl": 0.00634765625,
      "learning_rate": 2.1739130434782607e-07,
      "loss": 0.0006,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 83.546875,
      "epoch": 0.7888198757763976,
      "grad_norm": 9.345684051513672,
      "kl": 0.008056640625,
      "learning_rate": 2.1118012422360247e-07,
      "loss": 0.0008,
      "reward": 1.296875,
      "reward_mean": 1.296875,
      "reward_std": 0.19044628739356995,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "advantages": 0.0,
      "completion_length": 85.84375,
      "epoch": 0.7950310559006211,
      "grad_norm": 0.22835175693035126,
      "kl": 0.0084228515625,
      "learning_rate": 2.0496894409937888e-07,
      "loss": 0.0008,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "advantages": 0.0,
      "completion_length": 81.828125,
      "epoch": 0.8012422360248447,
      "grad_norm": 2.44989275932312,
      "kl": 0.007171630859375,
      "learning_rate": 1.9875776397515526e-07,
      "loss": 0.0007,
      "reward": 1.5,
      "reward_mean": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "advantages": 9.313225746154785e-10,
      "completion_length": 83.296875,
      "epoch": 0.8074534161490683,
      "grad_norm": 26.60379409790039,
      "kl": 0.01031494140625,
      "learning_rate": 1.9254658385093166e-07,
      "loss": 0.001,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.1530819982290268,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 75.296875,
      "epoch": 0.8136645962732919,
      "grad_norm": 2.649775981903076,
      "kl": 0.00787353515625,
      "learning_rate": 1.8633540372670807e-07,
      "loss": 0.0008,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "advantages": -7.450580596923828e-09,
      "completion_length": 72.921875,
      "epoch": 0.8198757763975155,
      "grad_norm": 6.021523952484131,
      "kl": 0.017333984375,
      "learning_rate": 1.8012422360248447e-07,
      "loss": 0.0017,
      "reward": 1.546875,
      "reward_mean": 1.546875,
      "reward_std": 0.17358146607875824,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 80.203125,
      "epoch": 0.8260869565217391,
      "grad_norm": 5.850553035736084,
      "kl": 0.0118408203125,
      "learning_rate": 1.7391304347826085e-07,
      "loss": 0.0012,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.25726157426834106,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 90.234375,
      "epoch": 0.8322981366459627,
      "grad_norm": 9.700899124145508,
      "kl": 0.01422119140625,
      "learning_rate": 1.6770186335403728e-07,
      "loss": 0.0014,
      "reward": 1.78125,
      "reward_mean": 1.78125,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.953125,
      "step": 134
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 81.90625,
      "epoch": 0.8385093167701864,
      "grad_norm": 2.9975473880767822,
      "kl": 0.00909423828125,
      "learning_rate": 1.6149068322981366e-07,
      "loss": 0.0009,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 79.3125,
      "epoch": 0.84472049689441,
      "grad_norm": 4.324582099914551,
      "kl": 0.01165771484375,
      "learning_rate": 1.5527950310559004e-07,
      "loss": 0.0012,
      "reward": 1.84375,
      "reward_mean": 1.84375,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 0.984375,
      "step": 136
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 80.265625,
      "epoch": 0.8509316770186336,
      "grad_norm": 3.8911736011505127,
      "kl": 0.009521484375,
      "learning_rate": 1.4906832298136647e-07,
      "loss": 0.001,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "advantages": 0.0,
      "completion_length": 78.40625,
      "epoch": 0.8571428571428571,
      "grad_norm": 2.864941120147705,
      "kl": 0.01129150390625,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0011,
      "reward": 1.765625,
      "reward_mean": 1.765625,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 138
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 80.21875,
      "epoch": 0.8633540372670807,
      "grad_norm": 5.788990497589111,
      "kl": 0.00799560546875,
      "learning_rate": 1.3664596273291925e-07,
      "loss": 0.0008,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 92.90625,
      "epoch": 0.8695652173913043,
      "grad_norm": 4.130926609039307,
      "kl": 0.0111083984375,
      "learning_rate": 1.3043478260869563e-07,
      "loss": 0.0011,
      "reward": 1.859375,
      "reward_mean": 1.859375,
      "reward_std": 0.2198973000049591,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 79.109375,
      "epoch": 0.8757763975155279,
      "grad_norm": 3.025212287902832,
      "kl": 0.01324462890625,
      "learning_rate": 1.2422360248447204e-07,
      "loss": 0.0013,
      "reward": 1.46875,
      "reward_mean": 1.46875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.984375,
      "step": 141
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 78.25,
      "epoch": 0.8819875776397516,
      "grad_norm": 6.828762531280518,
      "kl": 0.01177978515625,
      "learning_rate": 1.1801242236024844e-07,
      "loss": 0.0012,
      "reward": 1.40625,
      "reward_mean": 1.40625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "advantages": 0.0,
      "completion_length": 73.625,
      "epoch": 0.8881987577639752,
      "grad_norm": 3.4486515522003174,
      "kl": 0.00762939453125,
      "learning_rate": 1.1180124223602484e-07,
      "loss": 0.0008,
      "reward": 1.65625,
      "reward_mean": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 80.359375,
      "epoch": 0.8944099378881988,
      "grad_norm": 8.272978782653809,
      "kl": 0.00628662109375,
      "learning_rate": 1.0559006211180124e-07,
      "loss": 0.0006,
      "reward": 1.71875,
      "reward_mean": 1.71875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.984375,
      "step": 144
    },
    {
      "advantages": -7.450580596923828e-09,
      "completion_length": 81.1875,
      "epoch": 0.9006211180124224,
      "grad_norm": 4.848587512969971,
      "kl": 0.017578125,
      "learning_rate": 9.937888198757763e-08,
      "loss": 0.0018,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "advantages": 0.0,
      "completion_length": 79.984375,
      "epoch": 0.906832298136646,
      "grad_norm": 0.3604845702648163,
      "kl": 0.00958251953125,
      "learning_rate": 9.316770186335403e-08,
      "loss": 0.001,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 146
    },
    {
      "advantages": 5.587935447692871e-09,
      "completion_length": 76.5625,
      "epoch": 0.9130434782608695,
      "grad_norm": 10.680438995361328,
      "kl": 0.01116943359375,
      "learning_rate": 8.695652173913042e-08,
      "loss": 0.0011,
      "reward": 1.671875,
      "reward_mean": 1.671875,
      "reward_std": 0.19939783215522766,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 76.984375,
      "epoch": 0.9192546583850931,
      "grad_norm": 2.091907024383545,
      "kl": 0.010498046875,
      "learning_rate": 8.074534161490683e-08,
      "loss": 0.0011,
      "reward": 1.640625,
      "reward_mean": 1.640625,
      "reward_std": 0.08010874688625336,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.984375,
      "step": 148
    },
    {
      "advantages": 0.0,
      "completion_length": 76.203125,
      "epoch": 0.9254658385093167,
      "grad_norm": 0.20045147836208344,
      "kl": 0.0078125,
      "learning_rate": 7.453416149068323e-08,
      "loss": 0.0008,
      "reward": 1.75,
      "reward_mean": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "advantages": 0.0,
      "completion_length": 84.140625,
      "epoch": 0.9316770186335404,
      "grad_norm": 3.21720814704895,
      "kl": 0.008544921875,
      "learning_rate": 6.832298136645963e-08,
      "loss": 0.0009,
      "reward": 1.6875,
      "reward_mean": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 82.734375,
      "epoch": 0.937888198757764,
      "grad_norm": 7.955801963806152,
      "kl": 0.01263427734375,
      "learning_rate": 6.211180124223602e-08,
      "loss": 0.0013,
      "reward": 1.734375,
      "reward_mean": 1.734375,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 72.96875,
      "epoch": 0.9440993788819876,
      "grad_norm": 3.563530921936035,
      "kl": 0.0093994140625,
      "learning_rate": 5.590062111801242e-08,
      "loss": 0.0009,
      "reward": 1.90625,
      "reward_mean": 1.90625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 1.0,
      "step": 152
    },
    {
      "advantages": -9.313225746154785e-09,
      "completion_length": 83.125,
      "epoch": 0.9503105590062112,
      "grad_norm": 9.811988830566406,
      "kl": 0.0184326171875,
      "learning_rate": 4.9689440993788814e-08,
      "loss": 0.0019,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 153
    },
    {
      "advantages": -1.0244548320770264e-08,
      "completion_length": 83.515625,
      "epoch": 0.9565217391304348,
      "grad_norm": 3.8269639015197754,
      "kl": 0.0133056640625,
      "learning_rate": 4.347826086956521e-08,
      "loss": 0.0013,
      "reward": 1.78125,
      "reward_mean": 1.78125,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 154
    },
    {
      "advantages": -7.450580596923828e-09,
      "completion_length": 85.671875,
      "epoch": 0.9627329192546584,
      "grad_norm": 3.470165252685547,
      "kl": 0.01300048828125,
      "learning_rate": 3.726708074534162e-08,
      "loss": 0.0013,
      "reward": 1.5625,
      "reward_mean": 1.5625,
      "reward_std": 0.1462520956993103,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 155
    },
    {
      "advantages": -9.313225746154785e-10,
      "completion_length": 84.203125,
      "epoch": 0.968944099378882,
      "grad_norm": 2.550407648086548,
      "kl": 0.00946044921875,
      "learning_rate": 3.105590062111801e-08,
      "loss": 0.0009,
      "reward": 1.625,
      "reward_mean": 1.625,
      "reward_std": 0.16675157845020294,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.984375,
      "step": 156
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 76.296875,
      "epoch": 0.9751552795031055,
      "grad_norm": 3.396425247192383,
      "kl": 0.00714111328125,
      "learning_rate": 2.4844720496894407e-08,
      "loss": 0.0007,
      "reward": 1.546875,
      "reward_mean": 1.546875,
      "reward_std": 0.15992169082164764,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 157
    },
    {
      "advantages": -2.7939677238464355e-09,
      "completion_length": 73.859375,
      "epoch": 0.9813664596273292,
      "grad_norm": 3.776041030883789,
      "kl": 0.00921630859375,
      "learning_rate": 1.863354037267081e-08,
      "loss": 0.0009,
      "reward": 1.59375,
      "reward_mean": 1.59375,
      "reward_std": 0.10888782143592834,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 158
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 77.953125,
      "epoch": 0.9875776397515528,
      "grad_norm": 3.304471254348755,
      "kl": 0.01263427734375,
      "learning_rate": 1.2422360248447204e-08,
      "loss": 0.0013,
      "reward": 1.796875,
      "reward_mean": 1.796875,
      "reward_std": 0.11100947856903076,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "advantages": 0.0,
      "completion_length": 79.25,
      "epoch": 0.9937888198757764,
      "grad_norm": 5.823967456817627,
      "kl": 0.00897216796875,
      "learning_rate": 6.211180124223602e-09,
      "loss": 0.0009,
      "reward": 1.5,
      "reward_mean": 1.5,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "advantages": -0.5890890955924988,
      "completion_length": 89.33333587646484,
      "epoch": 1.0,
      "grad_norm": 2.0931286811828613,
      "kl": 0.00677490234375,
      "learning_rate": 0.0,
      "loss": 0.001,
      "reward": 1.6666667461395264,
      "reward_mean": 1.875,
      "reward_std": 0.3535533845424652,
      "rewards/accuracy_reward": 0.6666666865348816,
      "rewards/format_reward": 1.0,
      "step": 161
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 161,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}