{
  "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": 9.550005672210212,
      "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": 7.438416623721416,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -2.768181800842285,
      "logits/rejected": -2.7385101318359375,
      "logps/chosen": -256.52105712890625,
      "logps/pi_response": -72.76220703125,
      "logps/ref_response": -72.56087493896484,
      "logps/rejected": -173.57135009765625,
      "loss": 0.6872,
      "rewards/accuracies": 0.5902777910232544,
      "rewards/chosen": 0.022322500124573708,
      "rewards/margins": 0.01613031141459942,
      "rewards/rejected": 0.006192188244313002,
      "step": 10
    },
    {
      "epoch": 0.5095541401273885,
      "grad_norm": 6.100802491500333,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -2.7012898921966553,
      "logits/rejected": -2.676642894744873,
      "logps/chosen": -233.33203125,
      "logps/pi_response": -80.94047546386719,
      "logps/ref_response": -70.75907897949219,
      "logps/rejected": -180.34707641601562,
      "loss": 0.6566,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": 0.036385610699653625,
      "rewards/margins": 0.07418035715818405,
      "rewards/rejected": -0.037794746458530426,
      "step": 20
    },
    {
      "epoch": 0.7643312101910829,
      "grad_norm": 7.101547910056086,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -2.692800998687744,
      "logits/rejected": -2.6492133140563965,
      "logps/chosen": -257.54425048828125,
      "logps/pi_response": -101.40791320800781,
      "logps/ref_response": -70.76898193359375,
      "logps/rejected": -196.77926635742188,
      "loss": 0.6254,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.05073127895593643,
      "rewards/margins": 0.17368993163108826,
      "rewards/rejected": -0.22442121803760529,
      "step": 30
    },
    {
      "epoch": 0.9936305732484076,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 0.647348954127385,
      "train_runtime": 1784.9796,
      "train_samples_per_second": 5.602,
      "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
}