{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.587443946188341,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35874439461883406,
      "grad_norm": 5.573123455047607,
      "learning_rate": 9.259259259259259e-07,
      "logits/chosen": -2.3576977252960205,
      "logits/rejected": -2.3553850650787354,
      "logps/chosen": -61.0846061706543,
      "logps/rejected": -126.1152572631836,
      "loss": 0.6942,
      "rewards/accuracies": 0.36250001192092896,
      "rewards/chosen": -0.0012381849810481071,
      "rewards/margins": -0.010831715539097786,
      "rewards/rejected": 0.009593529626727104,
      "step": 10
    },
    {
      "epoch": 0.7174887892376681,
      "grad_norm": 3.8175244331359863,
      "learning_rate": 1.8518518518518519e-06,
      "logits/chosen": -2.3618054389953613,
      "logits/rejected": -2.384873867034912,
      "logps/chosen": -63.2865104675293,
      "logps/rejected": -81.8724365234375,
      "loss": 0.6925,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.002874236088246107,
      "rewards/margins": -0.003056168556213379,
      "rewards/rejected": 0.00018193255527876318,
      "step": 20
    },
    {
      "epoch": 1.0762331838565022,
      "grad_norm": 4.383662700653076,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": -2.3333091735839844,
      "logits/rejected": -2.3545360565185547,
      "logps/chosen": -76.02193450927734,
      "logps/rejected": -88.19218444824219,
      "loss": 0.6932,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.008687756024301052,
      "rewards/margins": -0.0016246589366346598,
      "rewards/rejected": -0.007063096854835749,
      "step": 30
    },
    {
      "epoch": 1.4349775784753362,
      "grad_norm": 4.095139503479004,
      "learning_rate": 3.7037037037037037e-06,
      "logits/chosen": -2.3483197689056396,
      "logits/rejected": -2.363131523132324,
      "logps/chosen": -66.08761596679688,
      "logps/rejected": -78.22222900390625,
      "loss": 0.6889,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.014177674427628517,
      "rewards/margins": 0.012472175993025303,
      "rewards/rejected": -0.026649847626686096,
      "step": 40
    },
    {
      "epoch": 1.7937219730941703,
      "grad_norm": 4.034746170043945,
      "learning_rate": 4.62962962962963e-06,
      "logits/chosen": -2.31459903717041,
      "logits/rejected": -2.335706949234009,
      "logps/chosen": -69.31814575195312,
      "logps/rejected": -77.91527557373047,
      "loss": 0.6823,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.04294499754905701,
      "rewards/margins": 0.007987136952579021,
      "rewards/rejected": -0.050932131707668304,
      "step": 50
    },
    {
      "epoch": 2.1524663677130045,
      "grad_norm": 4.620233535766602,
      "learning_rate": 4.998119881260576e-06,
      "logits/chosen": -2.364583969116211,
      "logits/rejected": -2.3598217964172363,
      "logps/chosen": -69.84207153320312,
      "logps/rejected": -79.67817687988281,
      "loss": 0.6717,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.06739393621683121,
      "rewards/margins": 0.050740111619234085,
      "rewards/rejected": -0.118134044110775,
      "step": 60
    },
    {
      "epoch": 2.5112107623318387,
      "grad_norm": 4.450082302093506,
      "learning_rate": 4.9866405060165044e-06,
      "logits/chosen": -2.298379421234131,
      "logits/rejected": -2.3169469833374023,
      "logps/chosen": -79.13768005371094,
      "logps/rejected": -79.24192810058594,
      "loss": 0.6499,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.1282506138086319,
      "rewards/margins": 0.08815944194793701,
      "rewards/rejected": -0.21641004085540771,
      "step": 70
    },
    {
      "epoch": 2.8699551569506725,
      "grad_norm": 4.7328877449035645,
      "learning_rate": 4.964774158361991e-06,
      "logits/chosen": -2.3363795280456543,
      "logits/rejected": -2.3331849575042725,
      "logps/chosen": -77.52214050292969,
      "logps/rejected": -86.90077209472656,
      "loss": 0.6346,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.22365322709083557,
      "rewards/margins": 0.1937221735715866,
      "rewards/rejected": -0.4173754155635834,
      "step": 80
    },
    {
      "epoch": 3.2286995515695067,
      "grad_norm": 4.016219615936279,
      "learning_rate": 4.93261217644956e-06,
      "logits/chosen": -2.30871844291687,
      "logits/rejected": -2.3049044609069824,
      "logps/chosen": -65.6957778930664,
      "logps/rejected": -77.02860260009766,
      "loss": 0.6004,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.23905189335346222,
      "rewards/margins": 0.2932591438293457,
      "rewards/rejected": -0.5323110222816467,
      "step": 90
    },
    {
      "epoch": 3.587443946188341,
      "grad_norm": 4.553714752197266,
      "learning_rate": 4.8902889044347e-06,
      "logits/chosen": -2.3068125247955322,
      "logits/rejected": -2.294462203979492,
      "logps/chosen": -85.46375274658203,
      "logps/rejected": -79.72854614257812,
      "loss": 0.5816,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.3506511151790619,
      "rewards/margins": 0.2997886538505554,
      "rewards/rejected": -0.6504397392272949,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.648437622595912e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}