{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.3966244725738397,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33755274261603374,
      "grad_norm": 0.4396141469478607,
      "learning_rate": 8.620689655172415e-07,
      "logits/chosen": 1.6453087329864502,
      "logits/rejected": 1.694819450378418,
      "logps/chosen": -74.5937728881836,
      "logps/rejected": -83.19783782958984,
      "loss": 0.6928,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": 0.003385844174772501,
      "rewards/margins": 0.004794469103217125,
      "rewards/rejected": -0.0014086246956139803,
      "step": 10
    },
    {
      "epoch": 0.6751054852320675,
      "grad_norm": 3.18404483795166,
      "learning_rate": 1.724137931034483e-06,
      "logits/chosen": 1.7945035696029663,
      "logits/rejected": 1.8347476720809937,
      "logps/chosen": -95.46636199951172,
      "logps/rejected": -101.22709655761719,
      "loss": 0.6933,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -0.0005593777750618756,
      "rewards/margins": -0.002013001125305891,
      "rewards/rejected": 0.0014536241069436073,
      "step": 20
    },
    {
      "epoch": 1.0126582278481013,
      "grad_norm": 0.48596081137657166,
      "learning_rate": 2.5862068965517246e-06,
      "logits/chosen": 1.774155616760254,
      "logits/rejected": 1.8374344110488892,
      "logps/chosen": -82.01265716552734,
      "logps/rejected": -84.23635864257812,
      "loss": 0.6948,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.0031326995231211185,
      "rewards/margins": -0.00463916826993227,
      "rewards/rejected": 0.0015064675826579332,
      "step": 30
    },
    {
      "epoch": 1.350210970464135,
      "grad_norm": 0.5144609808921814,
      "learning_rate": 3.448275862068966e-06,
      "logits/chosen": 1.838865876197815,
      "logits/rejected": 1.9505430459976196,
      "logps/chosen": -73.50071716308594,
      "logps/rejected": -88.8648452758789,
      "loss": 0.6928,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.0035957477521151304,
      "rewards/margins": -0.008830643258988857,
      "rewards/rejected": 0.005234894808381796,
      "step": 40
    },
    {
      "epoch": 1.6877637130801688,
      "grad_norm": 0.48295095562934875,
      "learning_rate": 4.310344827586207e-06,
      "logits/chosen": 1.8197529315948486,
      "logits/rejected": 1.8459796905517578,
      "logps/chosen": -83.36590576171875,
      "logps/rejected": -70.26930236816406,
      "loss": 0.6929,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.0030026868917047977,
      "rewards/margins": -0.006082554347813129,
      "rewards/rejected": 0.003079867223277688,
      "step": 50
    },
    {
      "epoch": 2.0464135021097047,
      "grad_norm": 0.5616092085838318,
      "learning_rate": 4.999818897894192e-06,
      "logits/chosen": 1.7525272369384766,
      "logits/rejected": 1.8092533349990845,
      "logps/chosen": -82.01802062988281,
      "logps/rejected": -78.39723205566406,
      "loss": 0.6949,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.0010920714121311903,
      "rewards/margins": -0.0016235255170613527,
      "rewards/rejected": 0.0005314538720995188,
      "step": 60
    },
    {
      "epoch": 2.3839662447257384,
      "grad_norm": 0.5237122774124146,
      "learning_rate": 4.9934830787948756e-06,
      "logits/chosen": 1.754880666732788,
      "logits/rejected": 1.8100305795669556,
      "logps/chosen": -89.62864685058594,
      "logps/rejected": -88.30036926269531,
      "loss": 0.694,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.002314353361725807,
      "rewards/margins": -0.001038494287058711,
      "rewards/rejected": -0.0012758590746670961,
      "step": 70
    },
    {
      "epoch": 2.721518987341772,
      "grad_norm": 0.507128119468689,
      "learning_rate": 4.978118375700895e-06,
      "logits/chosen": 1.7501789331436157,
      "logits/rejected": 1.6666221618652344,
      "logps/chosen": -94.74919128417969,
      "logps/rejected": -74.9794692993164,
      "loss": 0.6922,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": 0.00475720502436161,
      "rewards/margins": 0.01054429542273283,
      "rewards/rejected": -0.005787090864032507,
      "step": 80
    },
    {
      "epoch": 3.059071729957806,
      "grad_norm": 0.8151038885116577,
      "learning_rate": 4.953780424089803e-06,
      "logits/chosen": 1.7203710079193115,
      "logits/rejected": 1.7376699447631836,
      "logps/chosen": -76.8824691772461,
      "logps/rejected": -84.74555206298828,
      "loss": 0.6925,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": 0.01064519863575697,
      "rewards/margins": 0.012716365046799183,
      "rewards/rejected": -0.0020711664110422134,
      "step": 90
    },
    {
      "epoch": 3.3966244725738397,
      "grad_norm": 0.5476706027984619,
      "learning_rate": 4.920557351506409e-06,
      "logits/chosen": 1.7857105731964111,
      "logits/rejected": 1.7341340780258179,
      "logps/chosen": -88.39775085449219,
      "logps/rejected": -80.00445556640625,
      "loss": 0.6893,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.0052363635040819645,
      "rewards/margins": 0.016881367191672325,
      "rewards/rejected": -0.011645001359283924,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 580,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.428962066466079e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}