{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9921671018276762,
  "eval_steps": 100,
  "global_step": 95,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "logits/chosen": 0.8826487064361572,
      "logits/rejected": 0.921362042427063,
      "logps/chosen": -36.58121871948242,
      "logps/rejected": -54.902320861816406,
      "loss": 0.01,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": 0.8919395804405212,
      "logits/rejected": 0.8742702007293701,
      "logps/chosen": -87.81292724609375,
      "logps/rejected": -96.36103057861328,
      "loss": 0.0101,
      "rewards/accuracies": 0.1944444477558136,
      "rewards/chosen": -0.0001089610013877973,
      "rewards/margins": -0.00020776402379851788,
      "rewards/rejected": 9.880305151455104e-05,
      "step": 10
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.83118057351089e-06,
      "logits/chosen": 0.7794350385665894,
      "logits/rejected": 0.846326470375061,
      "logps/chosen": -91.74535369873047,
      "logps/rejected": -85.1441879272461,
      "loss": 0.01,
      "rewards/accuracies": 0.26249998807907104,
      "rewards/chosen": 0.0001209847250720486,
      "rewards/margins": 0.00020004021644126624,
      "rewards/rejected": -7.905550592113286e-05,
      "step": 20
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3475222930516484e-06,
      "logits/chosen": 0.86492520570755,
      "logits/rejected": 0.8484649658203125,
      "logps/chosen": -85.28953552246094,
      "logps/rejected": -78.07523345947266,
      "loss": 0.01,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.00029723229818046093,
      "rewards/margins": 0.0006696944474242628,
      "rewards/rejected": -0.00037246214924380183,
      "step": 30
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.6143458894413463e-06,
      "logits/chosen": 0.7646722793579102,
      "logits/rejected": 0.8417412042617798,
      "logps/chosen": -122.7321548461914,
      "logps/rejected": -108.7701187133789,
      "loss": 0.01,
      "rewards/accuracies": 0.29374998807907104,
      "rewards/chosen": 3.807881057582563e-06,
      "rewards/margins": -2.6598700060276315e-05,
      "rewards/rejected": 3.0406610676436685e-05,
      "step": 40
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.730670898658255e-06,
      "logits/chosen": 0.8404987454414368,
      "logits/rejected": 0.8694361448287964,
      "logps/chosen": -65.73135375976562,
      "logps/rejected": -71.68798828125,
      "loss": 0.01,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.0002373962925048545,
      "rewards/margins": 0.00048616877757012844,
      "rewards/rejected": -0.00024877244140952826,
      "step": 50
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8158425248197931e-06,
      "logits/chosen": 0.8144673109054565,
      "logits/rejected": 0.8853918313980103,
      "logps/chosen": -117.61384582519531,
      "logps/rejected": -110.96458435058594,
      "loss": 0.01,
      "rewards/accuracies": 0.3062500059604645,
      "rewards/chosen": 0.00019182416144758463,
      "rewards/margins": 0.0005247507360763848,
      "rewards/rejected": -0.00033292657462880015,
      "step": 60
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.934134090518593e-07,
      "logits/chosen": 0.8051176071166992,
      "logits/rejected": 0.8623756170272827,
      "logps/chosen": -82.68182373046875,
      "logps/rejected": -83.90242004394531,
      "loss": 0.01,
      "rewards/accuracies": 0.29374998807907104,
      "rewards/chosen": -0.00018608721438795328,
      "rewards/margins": -0.0002419232187094167,
      "rewards/rejected": 5.583601159742102e-05,
      "step": 70
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7445716067596506e-07,
      "logits/chosen": 0.7653516530990601,
      "logits/rejected": 0.8296685218811035,
      "logps/chosen": -105.7895736694336,
      "logps/rejected": -104.9858627319336,
      "loss": 0.01,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 0.00024473838857375085,
      "rewards/margins": 0.00043704017298296094,
      "rewards/rejected": -0.00019230166799388826,
      "step": 80
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.256725079024554e-08,
      "logits/chosen": 0.8357099294662476,
      "logits/rejected": 0.8563612103462219,
      "logps/chosen": -77.6849594116211,
      "logps/rejected": -90.00062561035156,
      "loss": 0.01,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 6.526726792799309e-05,
      "rewards/margins": 0.00044493190944194794,
      "rewards/rejected": -0.00037966464878991246,
      "step": 90
    },
    {
      "epoch": 0.99,
      "step": 95,
      "total_flos": 0.0,
      "train_loss": 0.009997876410029437,
      "train_runtime": 1153.3315,
      "train_samples_per_second": 5.3,
      "train_steps_per_second": 0.082
    }
  ],
  "logging_steps": 10,
  "max_steps": 95,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}