{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9936305732484076,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025477707006369428,
      "grad_norm": 9.541383295467487,
      "learning_rate": 1.25e-07,
      "logits/chosen": -2.9487781524658203,
      "logits/rejected": -2.858518123626709,
      "logps/chosen": -304.58709716796875,
      "logps/pi_response": -66.34120178222656,
      "logps/ref_response": -66.34120178222656,
      "logps/rejected": -133.11962890625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.25477707006369427,
      "grad_norm": 7.429726997596217,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -2.768242120742798,
      "logits/rejected": -2.7385332584381104,
      "logps/chosen": -256.5203552246094,
      "logps/pi_response": -72.73097229003906,
      "logps/ref_response": -72.56087493896484,
      "logps/rejected": -173.52818298339844,
      "loss": 0.6873,
      "rewards/accuracies": 0.6111111044883728,
      "rewards/chosen": 0.022329598665237427,
      "rewards/margins": 0.015705592930316925,
      "rewards/rejected": 0.006624007597565651,
      "step": 10
    },
    {
      "epoch": 0.5095541401273885,
      "grad_norm": 6.1021229114240585,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -2.701791524887085,
      "logits/rejected": -2.6771979331970215,
      "logps/chosen": -233.2255096435547,
      "logps/pi_response": -80.7745132446289,
      "logps/ref_response": -70.75907897949219,
      "logps/rejected": -180.2314453125,
      "loss": 0.6568,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": 0.03745081648230553,
      "rewards/margins": 0.07408934831619263,
      "rewards/rejected": -0.0366385318338871,
      "step": 20
    },
    {
      "epoch": 0.7643312101910829,
      "grad_norm": 7.0802381669561125,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -2.6936070919036865,
      "logits/rejected": -2.6500821113586426,
      "logps/chosen": -257.2088623046875,
      "logps/pi_response": -100.9505615234375,
      "logps/ref_response": -70.76898193359375,
      "logps/rejected": -196.41236877441406,
      "loss": 0.6258,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.047377265989780426,
      "rewards/margins": 0.17337492108345032,
      "rewards/rejected": -0.22075219452381134,
      "step": 30
    },
    {
      "epoch": 0.9936305732484076,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 0.6476312172718537,
      "train_runtime": 1765.539,
      "train_samples_per_second": 5.664,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}