{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.056338028169014086,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 480.421875,
      "epoch": 0.0028169014084507044,
      "grad_norm": 0.00048306785174645483,
      "learning_rate": 6.953333333333332e-07,
      "loss": 0.0002,
      "reward": 0.328125,
      "rewards/accuracy_reward": 0.328125,
      "rewards/tag_count_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 507.015625,
      "epoch": 0.005633802816901409,
      "grad_norm": 0.00925475638359785,
      "learning_rate": 6.906666666666666e-07,
      "loss": -0.0003,
      "reward": 0.265625,
      "rewards/accuracy_reward": 0.265625,
      "rewards/tag_count_reward": 0.0,
      "step": 2
    },
    {
      "completion_length": 452.828125,
      "epoch": 0.011267605633802818,
      "grad_norm": 0.0011863944819197059,
      "learning_rate": 6.813333333333333e-07,
      "loss": -0.0004,
      "reward": 0.4140625,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/tag_count_reward": 0.0,
      "step": 4
    },
    {
      "completion_length": 467.625,
      "epoch": 0.016901408450704224,
      "grad_norm": 0.00895125325769186,
      "learning_rate": 6.72e-07,
      "loss": -0.0,
      "reward": 0.3125,
      "rewards/accuracy_reward": 0.3125,
      "rewards/tag_count_reward": 0.0,
      "step": 6
    },
    {
      "completion_length": 507.421875,
      "epoch": 0.022535211267605635,
      "grad_norm": 0.00034816659172065556,
      "learning_rate": 6.626666666666666e-07,
      "loss": -0.0001,
      "reward": 0.359375,
      "rewards/accuracy_reward": 0.359375,
      "rewards/tag_count_reward": 0.0,
      "step": 8
    },
    {
      "completion_length": 497.6875,
      "epoch": 0.028169014084507043,
      "grad_norm": 0.033418115228414536,
      "learning_rate": 6.533333333333334e-07,
      "loss": -0.0002,
      "reward": 0.2109375,
      "rewards/accuracy_reward": 0.2109375,
      "rewards/tag_count_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 475.71875,
      "epoch": 0.03380281690140845,
      "grad_norm": 0.0005127754993736744,
      "learning_rate": 6.44e-07,
      "loss": -0.0003,
      "reward": 0.2734375,
      "rewards/accuracy_reward": 0.2734375,
      "rewards/tag_count_reward": 0.0,
      "step": 12
    },
    {
      "completion_length": 465.9765625,
      "epoch": 0.03943661971830986,
      "grad_norm": 0.001344753778539598,
      "learning_rate": 6.346666666666666e-07,
      "loss": -0.0007,
      "reward": 0.3203125,
      "rewards/accuracy_reward": 0.3203125,
      "rewards/tag_count_reward": 0.0,
      "step": 14
    },
    {
      "completion_length": 510.34375,
      "epoch": 0.04507042253521127,
      "grad_norm": 0.23263047635555267,
      "learning_rate": 6.253333333333333e-07,
      "loss": -0.0011,
      "reward": 0.4765625,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/tag_count_reward": 0.0,
      "step": 16
    },
    {
      "completion_length": 498.1484375,
      "epoch": 0.05070422535211268,
      "grad_norm": 0.0012522597098723054,
      "learning_rate": 6.16e-07,
      "loss": -0.0004,
      "reward": 0.2734375,
      "rewards/accuracy_reward": 0.2734375,
      "rewards/tag_count_reward": 0.0,
      "step": 18
    },
    {
      "completion_length": 509.1171875,
      "epoch": 0.056338028169014086,
      "grad_norm": 0.010728111490607262,
      "learning_rate": 6.066666666666666e-07,
      "loss": -0.0001,
      "reward": 0.328125,
      "rewards/accuracy_reward": 0.328125,
      "rewards/tag_count_reward": 0.0,
      "step": 20
    }
  ],
  "logging_steps": 2,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.832621368030003e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}