{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 313,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003194888178913738,
      "grad_norm": 40.63113754109814,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -0.72265625,
      "logits/rejected": -0.40234375,
      "logps/chosen": -296.0,
      "logps/rejected": -97.5,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 40.83376065165844,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.52734375,
      "logps/chosen": -326.0,
      "logps/rejected": -112.0,
      "loss": 0.6923,
      "rewards/accuracies": 0.1944444477558136,
      "rewards/chosen": -0.00592041015625,
      "rewards/margins": -0.0062255859375,
      "rewards/rejected": 0.0003452301025390625,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 36.83108395610394,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -0.75390625,
      "logits/rejected": -0.6171875,
      "logps/chosen": -360.0,
      "logps/rejected": -116.5,
      "loss": 0.6765,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.00433349609375,
      "rewards/margins": 0.0306396484375,
      "rewards/rejected": -0.03515625,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 33.32858154831051,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -0.8515625,
      "logits/rejected": -0.6015625,
      "logps/chosen": -342.0,
      "logps/rejected": -124.0,
      "loss": 0.6042,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 0.049072265625,
      "rewards/margins": 0.21484375,
      "rewards/rejected": -0.166015625,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 27.32264489173027,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -0.70703125,
      "logits/rejected": -0.58984375,
      "logps/chosen": -356.0,
      "logps/rejected": -124.0,
      "loss": 0.4774,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.12451171875,
      "rewards/margins": 0.5859375,
      "rewards/rejected": -0.4609375,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 15.975644864718246,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -0.79296875,
      "logits/rejected": -0.5546875,
      "logps/chosen": -286.0,
      "logps/rejected": -126.5,
      "loss": 0.2903,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.05859375,
      "rewards/margins": 1.109375,
      "rewards/rejected": -1.046875,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 6.256507586550031,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.50390625,
      "logps/chosen": -302.0,
      "logps/rejected": -140.0,
      "loss": 0.129,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.056396484375,
      "rewards/margins": 2.328125,
      "rewards/rejected": -2.28125,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 3.6739087198855866,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -0.74609375,
      "logits/rejected": -0.435546875,
      "logps/chosen": -300.0,
      "logps/rejected": -152.0,
      "loss": 0.0324,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.197265625,
      "rewards/margins": 4.375,
      "rewards/rejected": -4.15625,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 5.267432309185282,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -0.76171875,
      "logits/rejected": -0.5,
      "logps/chosen": -296.0,
      "logps/rejected": -174.0,
      "loss": 0.0112,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.341796875,
      "rewards/margins": 5.9375,
      "rewards/rejected": -5.59375,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 2.308803456420075,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -0.60546875,
      "logits/rejected": -0.423828125,
      "logps/chosen": -255.0,
      "logps/rejected": -171.0,
      "loss": 0.006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.462890625,
      "rewards/margins": 6.75,
      "rewards/rejected": -6.28125,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.19555856015239148,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -0.7421875,
      "logits/rejected": -0.56640625,
      "logps/chosen": -356.0,
      "logps/rejected": -191.0,
      "loss": 0.0026,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.53125,
      "rewards/margins": 7.84375,
      "rewards/rejected": -7.3125,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.11747858932749748,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -0.6640625,
      "logits/rejected": -0.56640625,
      "logps/chosen": -294.0,
      "logps/rejected": -201.0,
      "loss": 0.0062,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.46484375,
      "rewards/margins": 8.25,
      "rewards/rejected": -7.75,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.2954350664468083,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -0.6015625,
      "logits/rejected": -0.373046875,
      "logps/chosen": -240.0,
      "logps/rejected": -189.0,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.380859375,
      "rewards/margins": 8.3125,
      "rewards/rejected": -7.9375,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.4447368308978073,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -0.578125,
      "logits/rejected": -0.54296875,
      "logps/chosen": -314.0,
      "logps/rejected": -205.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.42578125,
      "rewards/margins": 8.6875,
      "rewards/rejected": -8.25,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.13989545962415664,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -0.6484375,
      "logits/rejected": -0.494140625,
      "logps/chosen": -346.0,
      "logps/rejected": -203.0,
      "loss": 0.0025,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2158203125,
      "rewards/margins": 9.1875,
      "rewards/rejected": -9.0,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 1.005715548136931,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.5859375,
      "logps/chosen": -322.0,
      "logps/rejected": -206.0,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.466796875,
      "rewards/margins": 9.6875,
      "rewards/rejected": -9.1875,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.015257918360502878,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.447265625,
      "logps/chosen": -312.0,
      "logps/rejected": -216.0,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.125,
      "rewards/margins": 9.75,
      "rewards/rejected": -9.625,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.14450582552206107,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.5625,
      "logps/chosen": -276.0,
      "logps/rejected": -208.0,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.271484375,
      "rewards/margins": 9.9375,
      "rewards/rejected": -9.6875,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.04115746281274911,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.5625,
      "logps/chosen": -332.0,
      "logps/rejected": -217.0,
      "loss": 0.0083,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2470703125,
      "rewards/margins": 9.9375,
      "rewards/rejected": -9.6875,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.027254932142989203,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -0.578125,
      "logits/rejected": -0.48046875,
      "logps/chosen": -306.0,
      "logps/rejected": -219.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1982421875,
      "rewards/margins": 10.375,
      "rewards/rejected": -10.1875,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.04076702181699882,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -0.7265625,
      "logits/rejected": -0.478515625,
      "logps/chosen": -348.0,
      "logps/rejected": -226.0,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.125,
      "rewards/margins": 10.6875,
      "rewards/rejected": -10.5,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.02362186732429131,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -0.71875,
      "logits/rejected": -0.5,
      "logps/chosen": -316.0,
      "logps/rejected": -220.0,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.11328125,
      "rewards/margins": 10.375,
      "rewards/rejected": -10.5,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.019809535140906925,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -0.6796875,
      "logits/rejected": -0.57421875,
      "logps/chosen": -356.0,
      "logps/rejected": -228.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.01483154296875,
      "rewards/margins": 10.5625,
      "rewards/rejected": -10.5625,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 1.5883393714345544,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.5078125,
      "logps/chosen": -298.0,
      "logps/rejected": -220.0,
      "loss": 0.007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.0810546875,
      "rewards/margins": 10.6875,
      "rewards/rejected": -10.75,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.020553702943879617,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -0.62890625,
      "logits/rejected": -0.55859375,
      "logps/chosen": -336.0,
      "logps/rejected": -230.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.034423828125,
      "rewards/margins": 10.9375,
      "rewards/rejected": -10.9375,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.18543484770077023,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -0.75,
      "logits/rejected": -0.458984375,
      "logps/chosen": -330.0,
      "logps/rejected": -221.0,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2392578125,
      "rewards/margins": 10.4375,
      "rewards/rejected": -10.625,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.025956022709723227,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.59765625,
      "logps/chosen": -322.0,
      "logps/rejected": -237.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.326171875,
      "rewards/margins": 10.9375,
      "rewards/rejected": -11.3125,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.768139770920166,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -0.703125,
      "logits/rejected": -0.57421875,
      "logps/chosen": -304.0,
      "logps/rejected": -228.0,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.0089111328125,
      "rewards/margins": 11.4375,
      "rewards/rejected": -11.4375,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.5159058299026532,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -0.59765625,
      "logits/rejected": -0.46484375,
      "logps/chosen": -296.0,
      "logps/rejected": -230.0,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.458984375,
      "rewards/margins": 11.0625,
      "rewards/rejected": -11.5,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.0031165260658612332,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -0.76171875,
      "logits/rejected": -0.625,
      "logps/chosen": -338.0,
      "logps/rejected": -234.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.306640625,
      "rewards/margins": 11.375,
      "rewards/rejected": -11.6875,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.013736141734169007,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -0.69140625,
      "logits/rejected": -0.5078125,
      "logps/chosen": -328.0,
      "logps/rejected": -240.0,
      "loss": 0.0014,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.28125,
      "rewards/margins": 11.75,
      "rewards/rejected": -12.0625,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.031010940779375298,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -0.5390625,
      "logits/rejected": -0.515625,
      "logps/chosen": -300.0,
      "logps/rejected": -237.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.439453125,
      "rewards/margins": 11.75,
      "rewards/rejected": -12.1875,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -0.72265625,
      "eval_logits/rejected": -0.4375,
      "eval_logps/chosen": -296.0,
      "eval_logps/rejected": -233.0,
      "eval_loss": 0.0003721904649864882,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -0.291015625,
      "eval_rewards/margins": 11.625,
      "eval_rewards/rejected": -11.9375,
      "eval_runtime": 19.9934,
      "eval_samples_per_second": 10.003,
      "eval_steps_per_second": 0.35,
      "step": 313
    }
  ],
  "logging_steps": 10,
  "max_steps": 939,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}