{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9996261682242991,
  "eval_steps": 134,
  "global_step": 1337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05009345794392523,
      "grad_norm": 1.3687894344329834,
      "learning_rate": 5e-07,
      "logits/chosen": -0.6576648950576782,
      "logits/rejected": -0.6672008633613586,
      "logps/chosen": -2.227020502090454,
      "logps/rejected": -5.950113296508789,
      "loss": 0.6927,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.0005052318447269499,
      "rewards/margins": 0.0008711821283213794,
      "rewards/rejected": -0.0003659502835944295,
      "step": 67
    },
    {
      "epoch": 0.10018691588785046,
      "grad_norm": 2.2226035594940186,
      "learning_rate": 1e-06,
      "logits/chosen": -0.643012285232544,
      "logits/rejected": -0.6527656316757202,
      "logps/chosen": -2.1846628189086914,
      "logps/rejected": -6.394010066986084,
      "loss": 0.6883,
      "rewards/accuracies": 0.7276118993759155,
      "rewards/chosen": -0.0008213059627451003,
      "rewards/margins": 0.00972337368875742,
      "rewards/rejected": -0.01054468099027872,
      "step": 134
    },
    {
      "epoch": 0.1502803738317757,
      "grad_norm": 2.133711576461792,
      "learning_rate": 9.443059019118869e-07,
      "logits/chosen": -0.6600874662399292,
      "logits/rejected": -0.6701396703720093,
      "logps/chosen": -2.2199361324310303,
      "logps/rejected": -7.209847450256348,
      "loss": 0.6524,
      "rewards/accuracies": 0.8563432693481445,
      "rewards/chosen": -0.0032122363336384296,
      "rewards/margins": 0.08549854904413223,
      "rewards/rejected": -0.08871078491210938,
      "step": 201
    },
    {
      "epoch": 0.20037383177570092,
      "grad_norm": 3.0340538024902344,
      "learning_rate": 8.886118038237738e-07,
      "logits/chosen": -0.6653879284858704,
      "logits/rejected": -0.677828848361969,
      "logps/chosen": -2.533212184906006,
      "logps/rejected": -10.03713321685791,
      "loss": 0.5533,
      "rewards/accuracies": 0.8488805890083313,
      "rewards/chosen": -0.02541249990463257,
      "rewards/margins": 0.32844945788383484,
      "rewards/rejected": -0.3538619577884674,
      "step": 268
    },
    {
      "epoch": 0.2504672897196262,
      "grad_norm": 2.396055221557617,
      "learning_rate": 8.329177057356608e-07,
      "logits/chosen": -0.6770755052566528,
      "logits/rejected": -0.6937317848205566,
      "logps/chosen": -3.257599353790283,
      "logps/rejected": -14.972702980041504,
      "loss": 0.4397,
      "rewards/accuracies": 0.8526118993759155,
      "rewards/chosen": -0.10955464839935303,
      "rewards/margins": 0.7154061198234558,
      "rewards/rejected": -0.8249607682228088,
      "step": 335
    },
    {
      "epoch": 0.3005607476635514,
      "grad_norm": 2.1619019508361816,
      "learning_rate": 7.772236076475478e-07,
      "logits/chosen": -0.7039727568626404,
      "logits/rejected": -0.7222999334335327,
      "logps/chosen": -3.5498507022857666,
      "logps/rejected": -20.35832405090332,
      "loss": 0.3224,
      "rewards/accuracies": 0.9029850363731384,
      "rewards/chosen": -0.15284313261508942,
      "rewards/margins": 1.2502268552780151,
      "rewards/rejected": -1.4030699729919434,
      "step": 402
    },
    {
      "epoch": 0.3506542056074766,
      "grad_norm": 4.72946834564209,
      "learning_rate": 7.215295095594347e-07,
      "logits/chosen": -0.7181271910667419,
      "logits/rejected": -0.7374780774116516,
      "logps/chosen": -5.401639461517334,
      "logps/rejected": -25.78311538696289,
      "loss": 0.2782,
      "rewards/accuracies": 0.9048507213592529,
      "rewards/chosen": -0.3204249441623688,
      "rewards/margins": 1.6534833908081055,
      "rewards/rejected": -1.9739083051681519,
      "step": 469
    },
    {
      "epoch": 0.40074766355140184,
      "grad_norm": 0.9766824245452881,
      "learning_rate": 6.658354114713217e-07,
      "logits/chosen": -0.756517231464386,
      "logits/rejected": -0.7776155471801758,
      "logps/chosen": -5.15533971786499,
      "logps/rejected": -30.16547203063965,
      "loss": 0.2283,
      "rewards/accuracies": 0.9123134016990662,
      "rewards/chosen": -0.3061079978942871,
      "rewards/margins": 2.072856903076172,
      "rewards/rejected": -2.37896466255188,
      "step": 536
    },
    {
      "epoch": 0.4508411214953271,
      "grad_norm": 4.220199108123779,
      "learning_rate": 6.101413133832086e-07,
      "logits/chosen": -0.7682312726974487,
      "logits/rejected": -0.7901257872581482,
      "logps/chosen": -6.049707412719727,
      "logps/rejected": -31.82540512084961,
      "loss": 0.2276,
      "rewards/accuracies": 0.9048507213592529,
      "rewards/chosen": -0.37660378217697144,
      "rewards/margins": 2.2089908123016357,
      "rewards/rejected": -2.585594415664673,
      "step": 603
    },
    {
      "epoch": 0.5009345794392523,
      "grad_norm": 6.615551471710205,
      "learning_rate": 5.544472152950955e-07,
      "logits/chosen": -0.7774823904037476,
      "logits/rejected": -0.8002229332923889,
      "logps/chosen": -6.285848617553711,
      "logps/rejected": -35.35755157470703,
      "loss": 0.1959,
      "rewards/accuracies": 0.9197760820388794,
      "rewards/chosen": -0.40754997730255127,
      "rewards/margins": 2.4701759815216064,
      "rewards/rejected": -2.8777260780334473,
      "step": 670
    },
    {
      "epoch": 0.5510280373831775,
      "grad_norm": 3.6465256214141846,
      "learning_rate": 4.987531172069825e-07,
      "logits/chosen": -0.7746462821960449,
      "logits/rejected": -0.7996317148208618,
      "logps/chosen": -7.126999378204346,
      "logps/rejected": -37.35933303833008,
      "loss": 0.1813,
      "rewards/accuracies": 0.9309701323509216,
      "rewards/chosen": -0.4826628267765045,
      "rewards/margins": 2.6208343505859375,
      "rewards/rejected": -3.10349702835083,
      "step": 737
    },
    {
      "epoch": 0.6011214953271028,
      "grad_norm": 9.422818183898926,
      "learning_rate": 4.4305901911886947e-07,
      "logits/chosen": -0.7760910391807556,
      "logits/rejected": -0.802481472492218,
      "logps/chosen": -6.253903388977051,
      "logps/rejected": -37.6789436340332,
      "loss": 0.1769,
      "rewards/accuracies": 0.9402984976768494,
      "rewards/chosen": -0.3954426348209381,
      "rewards/margins": 2.7266414165496826,
      "rewards/rejected": -3.122084140777588,
      "step": 804
    },
    {
      "epoch": 0.6512149532710281,
      "grad_norm": 6.800887107849121,
      "learning_rate": 3.873649210307564e-07,
      "logits/chosen": -0.7794383764266968,
      "logits/rejected": -0.8046081066131592,
      "logps/chosen": -6.785355567932129,
      "logps/rejected": -38.823604583740234,
      "loss": 0.1737,
      "rewards/accuracies": 0.9291044473648071,
      "rewards/chosen": -0.43714043498039246,
      "rewards/margins": 2.833800792694092,
      "rewards/rejected": -3.2709412574768066,
      "step": 871
    },
    {
      "epoch": 0.7013084112149532,
      "grad_norm": 4.8703718185424805,
      "learning_rate": 3.3167082294264335e-07,
      "logits/chosen": -0.7899920344352722,
      "logits/rejected": -0.8165501952171326,
      "logps/chosen": -7.20438814163208,
      "logps/rejected": -39.9392204284668,
      "loss": 0.1891,
      "rewards/accuracies": 0.9272387623786926,
      "rewards/chosen": -0.47980982065200806,
      "rewards/margins": 2.8853275775909424,
      "rewards/rejected": -3.3651373386383057,
      "step": 938
    },
    {
      "epoch": 0.7514018691588785,
      "grad_norm": 0.2955164611339569,
      "learning_rate": 2.7597672485453034e-07,
      "logits/chosen": -0.7796534895896912,
      "logits/rejected": -0.8110276460647583,
      "logps/chosen": -4.945315361022949,
      "logps/rejected": -41.536590576171875,
      "loss": 0.1215,
      "rewards/accuracies": 0.9570895433425903,
      "rewards/chosen": -0.2934176027774811,
      "rewards/margins": 3.1789510250091553,
      "rewards/rejected": -3.4723687171936035,
      "step": 1005
    },
    {
      "epoch": 0.8014953271028037,
      "grad_norm": 6.8209075927734375,
      "learning_rate": 2.2028262676641728e-07,
      "logits/chosen": -0.7921628952026367,
      "logits/rejected": -0.8196142315864563,
      "logps/chosen": -6.230679512023926,
      "logps/rejected": -41.968841552734375,
      "loss": 0.1502,
      "rewards/accuracies": 0.9477611780166626,
      "rewards/chosen": -0.4138028919696808,
      "rewards/margins": 3.128687858581543,
      "rewards/rejected": -3.5424907207489014,
      "step": 1072
    },
    {
      "epoch": 0.851588785046729,
      "grad_norm": 6.143770217895508,
      "learning_rate": 1.6458852867830422e-07,
      "logits/chosen": -0.7991640567779541,
      "logits/rejected": -0.8256679773330688,
      "logps/chosen": -6.564340114593506,
      "logps/rejected": -42.35133361816406,
      "loss": 0.1459,
      "rewards/accuracies": 0.9402984976768494,
      "rewards/chosen": -0.44414621591567993,
      "rewards/margins": 3.182438611984253,
      "rewards/rejected": -3.626584529876709,
      "step": 1139
    },
    {
      "epoch": 0.9016822429906542,
      "grad_norm": 5.307474613189697,
      "learning_rate": 1.0889443059019118e-07,
      "logits/chosen": -0.7985506057739258,
      "logits/rejected": -0.8279980421066284,
      "logps/chosen": -5.217541217803955,
      "logps/rejected": -43.41652297973633,
      "loss": 0.1275,
      "rewards/accuracies": 0.9496268630027771,
      "rewards/chosen": -0.3293641209602356,
      "rewards/margins": 3.362889289855957,
      "rewards/rejected": -3.692253351211548,
      "step": 1206
    },
    {
      "epoch": 0.9517757009345794,
      "grad_norm": 5.2688984870910645,
      "learning_rate": 5.320033250207814e-08,
      "logits/chosen": -0.7938746809959412,
      "logits/rejected": -0.8232108354568481,
      "logps/chosen": -5.571805477142334,
      "logps/rejected": -43.23577880859375,
      "loss": 0.1313,
      "rewards/accuracies": 0.9589552283287048,
      "rewards/chosen": -0.33297136425971985,
      "rewards/margins": 3.3113536834716797,
      "rewards/rejected": -3.644325017929077,
      "step": 1273
    },
    {
      "epoch": 0.9996261682242991,
      "step": 1337,
      "total_flos": 1.8353312732676096e+17,
      "train_loss": 0.2906364963049307,
      "train_runtime": 4209.7283,
      "train_samples_per_second": 2.542,
      "train_steps_per_second": 0.318
    }
  ],
  "logging_steps": 67,
  "max_steps": 1337,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8353312732676096e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}