{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 100,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 17.039526624858368,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -16.87786293029785,
      "logits/rejected": -17.083940505981445,
      "logps/chosen": -334.51202392578125,
      "logps/rejected": -430.7342224121094,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 17.02641660622309,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -17.35299301147461,
      "logits/rejected": -17.494272232055664,
      "logps/chosen": -402.1957702636719,
      "logps/rejected": -436.8443603515625,
      "loss": 0.6942,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": 0.0008524802979081869,
      "rewards/margins": 0.0015829234616830945,
      "rewards/rejected": -0.000730442872736603,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 15.563346495094395,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": -16.698389053344727,
      "logits/rejected": -16.837512969970703,
      "logps/chosen": -379.29327392578125,
      "logps/rejected": -413.91607666015625,
      "loss": 0.689,
      "rewards/accuracies": 0.4312500059604645,
      "rewards/chosen": 0.007832124829292297,
      "rewards/margins": -0.00933709554374218,
      "rewards/rejected": 0.017169222235679626,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 17.011437827705663,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": -17.711877822875977,
      "logits/rejected": -17.14129638671875,
      "logps/chosen": -398.89691162109375,
      "logps/rejected": -415.79156494140625,
      "loss": 0.6725,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.08761711418628693,
      "rewards/margins": 0.03942377120256424,
      "rewards/rejected": 0.04819334298372269,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 14.678769570259123,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": -17.091754913330078,
      "logits/rejected": -16.351659774780273,
      "logps/chosen": -358.50946044921875,
      "logps/rejected": -351.41278076171875,
      "loss": 0.6555,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.22874267399311066,
      "rewards/margins": 0.08902247995138168,
      "rewards/rejected": 0.13972017168998718,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 16.97842458896973,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": -18.3736515045166,
      "logits/rejected": -18.453527450561523,
      "logps/chosen": -411.6908264160156,
      "logps/rejected": -464.92529296875,
      "loss": 0.6319,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": 0.4013705253601074,
      "rewards/margins": 0.18392089009284973,
      "rewards/rejected": 0.2174496352672577,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 14.380477446845697,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": -18.0875301361084,
      "logits/rejected": -17.782365798950195,
      "logps/chosen": -400.21282958984375,
      "logps/rejected": -429.13922119140625,
      "loss": 0.6034,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.38667935132980347,
      "rewards/margins": 0.23242299258708954,
      "rewards/rejected": 0.15425635874271393,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 13.517572081923934,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": -17.54964828491211,
      "logits/rejected": -17.380077362060547,
      "logps/chosen": -369.05426025390625,
      "logps/rejected": -405.892578125,
      "loss": 0.5985,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 0.3417219817638397,
      "rewards/margins": 0.23048743605613708,
      "rewards/rejected": 0.11123454570770264,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 13.094353030731684,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": -17.020626068115234,
      "logits/rejected": -17.626625061035156,
      "logps/chosen": -355.83404541015625,
      "logps/rejected": -411.6244201660156,
      "loss": 0.5786,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": 0.23768803477287292,
      "rewards/margins": 0.33307021856307983,
      "rewards/rejected": -0.09538215398788452,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 14.853001845607317,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": -17.405044555664062,
      "logits/rejected": -17.63189697265625,
      "logps/chosen": -387.94354248046875,
      "logps/rejected": -437.82366943359375,
      "loss": 0.5742,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": 0.27590411901474,
      "rewards/margins": 0.3552981913089752,
      "rewards/rejected": -0.07939404994249344,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 13.859137885848625,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": -17.7724552154541,
      "logits/rejected": -17.65437889099121,
      "logps/chosen": -380.60955810546875,
      "logps/rejected": -411.00164794921875,
      "loss": 0.5725,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 0.2799278199672699,
      "rewards/margins": 0.2864844799041748,
      "rewards/rejected": -0.006556662730872631,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": -16.585969924926758,
      "eval_logits/rejected": -15.948739051818848,
      "eval_logps/chosen": -363.741943359375,
      "eval_logps/rejected": -363.7130432128906,
      "eval_loss": 0.5959563255310059,
      "eval_rewards/accuracies": 0.71875,
      "eval_rewards/chosen": 0.22609315812587738,
      "eval_rewards/margins": 0.2597140967845917,
      "eval_rewards/rejected": -0.03362090513110161,
      "eval_runtime": 8.7004,
      "eval_samples_per_second": 86.203,
      "eval_steps_per_second": 2.758,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.6262502945386447,
      "train_runtime": 431.1142,
      "train_samples_per_second": 31.314,
      "train_steps_per_second": 0.241
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}