trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 1,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 5e-07,
      "logits/chosen": -2.8170723915100098,
      "logits/rejected": -2.8022234439849854,
      "logps/chosen": -362.7491455078125,
      "logps/rejected": -374.0892639160156,
      "loss": 0.3466,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 1.0,
      "step": 1,
      "total_flos": 0.0,
      "train_loss": 0.3465735912322998,
      "train_runtime": 27.0943,
      "train_samples_per_second": 1.107,
      "train_steps_per_second": 0.037
    }
  ],
  "logging_steps": 10,
  "max_steps": 1,
  "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
}