{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9997984277363435,
  "eval_steps": 155,
  "global_step": 310,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49989921386817177,
      "grad_norm": 47.42934799194336,
      "learning_rate": 2.6360544217687075e-07,
      "logits/chosen": 0.009249088354408741,
      "logits/rejected": 0.021735819056630135,
      "logps/chosen": -528.992431640625,
      "logps/rejected": -473.0443115234375,
      "loss": 0.6567,
      "rewards/accuracies": 0.5882056355476379,
      "rewards/chosen": 0.4344692528247833,
      "rewards/margins": 0.12380775809288025,
      "rewards/rejected": 0.31066152453422546,
      "step": 155
    },
    {
      "epoch": 0.49989921386817177,
      "eval_logits/chosen": 0.00047391001135110855,
      "eval_logits/rejected": 0.010226032696664333,
      "eval_logps/chosen": -530.94873046875,
      "eval_logps/rejected": -473.15032958984375,
      "eval_loss": 0.6275001168251038,
      "eval_rewards/accuracies": 0.6649543642997742,
      "eval_rewards/chosen": 0.6048734188079834,
      "eval_rewards/margins": 0.22375237941741943,
      "eval_rewards/rejected": 0.3811211884021759,
      "eval_runtime": 1606.8252,
      "eval_samples_per_second": 10.897,
      "eval_steps_per_second": 0.273,
      "step": 155
    },
    {
      "epoch": 0.9997984277363435,
      "grad_norm": 14.372474670410156,
      "learning_rate": 0.0,
      "logits/chosen": 0.003721719840541482,
      "logits/rejected": 0.0137369679287076,
      "logps/chosen": -528.1862182617188,
      "logps/rejected": -464.98052978515625,
      "loss": 0.614,
      "rewards/accuracies": 0.66401207447052,
      "rewards/chosen": 0.6172089576721191,
      "rewards/margins": 0.2706223726272583,
      "rewards/rejected": 0.34658658504486084,
      "step": 310
    },
    {
      "epoch": 0.9997984277363435,
      "eval_logits/chosen": -0.0003721735847648233,
      "eval_logits/rejected": 0.009112725965678692,
      "eval_logps/chosen": -530.7772827148438,
      "eval_logps/rejected": -473.510009765625,
      "eval_loss": 0.6093795895576477,
      "eval_rewards/accuracies": 0.6797945499420166,
      "eval_rewards/chosen": 0.6220178604125977,
      "eval_rewards/margins": 0.27686768770217896,
      "eval_rewards/rejected": 0.3451501727104187,
      "eval_runtime": 1606.0288,
      "eval_samples_per_second": 10.903,
      "eval_steps_per_second": 0.273,
      "step": 310
    }
  ],
  "logging_steps": 155,
  "max_steps": 310,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 155,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}