{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 128,
  "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": 206.83701472895862,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 117.53560638427734,
      "logits/rejected": 126.8960952758789,
      "logps/chosen": -335.40118408203125,
      "logps/rejected": -439.16552734375,
      "loss": 0.8466,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 143.02246273043633,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 135.25250244140625,
      "logits/rejected": 138.58753967285156,
      "logps/chosen": -394.8367919921875,
      "logps/rejected": -438.3687744140625,
      "loss": 0.8168,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": 0.05798730626702309,
      "rewards/margins": 0.010051175951957703,
      "rewards/rejected": 0.047936126589775085,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 123.44922716715021,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 123.2398681640625,
      "logits/rejected": 127.10523986816406,
      "logps/chosen": -359.848876953125,
      "logps/rejected": -408.9041748046875,
      "loss": 0.77,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": 0.6281569600105286,
      "rewards/margins": 0.07379511743783951,
      "rewards/rejected": 0.5543618202209473,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 191.25695289070478,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 141.13412475585938,
      "logits/rejected": 134.9121551513672,
      "logps/chosen": -413.17120361328125,
      "logps/rejected": -454.4007873535156,
      "loss": 0.6605,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.0352973937988281,
      "rewards/margins": 0.6846312284469604,
      "rewards/rejected": -1.7199283838272095,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 117.3229813343473,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 123.485107421875,
      "logits/rejected": 112.0093994140625,
      "logps/chosen": -407.3685302734375,
      "logps/rejected": -430.86260986328125,
      "loss": 0.6439,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -2.6587166786193848,
      "rewards/margins": 0.9906498193740845,
      "rewards/rejected": -3.6493663787841797,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 135.55553476426843,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 138.61500549316406,
      "logits/rejected": 142.8344268798828,
      "logps/chosen": -459.23187255859375,
      "logps/rejected": -541.0479125976562,
      "loss": 0.5715,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -2.0044033527374268,
      "rewards/margins": 1.2316728830337524,
      "rewards/rejected": -3.2360763549804688,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 84.07274179093439,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 130.79531860351562,
      "logits/rejected": 133.99810791015625,
      "logps/chosen": -439.33355712890625,
      "logps/rejected": -517.16943359375,
      "loss": 0.4547,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -2.0663931369781494,
      "rewards/margins": 1.973799467086792,
      "rewards/rejected": -4.0401930809021,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 73.9831562560362,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 123.79913330078125,
      "logits/rejected": 125.98814392089844,
      "logps/chosen": -418.53662109375,
      "logps/rejected": -511.00213623046875,
      "loss": 0.388,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -2.5696399211883545,
      "rewards/margins": 2.3047735691070557,
      "rewards/rejected": -4.87441349029541,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 71.02459820713865,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 110.6953125,
      "logits/rejected": 127.4843521118164,
      "logps/chosen": -417.51416015625,
      "logps/rejected": -526.156005859375,
      "loss": 0.364,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": -3.1931560039520264,
      "rewards/margins": 2.4527578353881836,
      "rewards/rejected": -5.645913124084473,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 93.3873120821413,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 111.47325134277344,
      "logits/rejected": 124.3362045288086,
      "logps/chosen": -446.8865661621094,
      "logps/rejected": -550.5967407226562,
      "loss": 0.3545,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -3.022777557373047,
      "rewards/margins": 2.622969627380371,
      "rewards/rejected": -5.645747184753418,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 74.82643608321943,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 118.45140075683594,
      "logits/rejected": 117.85823822021484,
      "logps/chosen": -434.776123046875,
      "logps/rejected": -516.4940795898438,
      "loss": 0.3633,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -2.6400508880615234,
      "rewards/margins": 2.3464527130126953,
      "rewards/rejected": -4.986503601074219,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.5405629426240921,
      "train_runtime": 2172.3856,
      "train_samples_per_second": 6.214,
      "train_steps_per_second": 0.048
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}