{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.861635220125786,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.3409,
      "eval_samples_per_second": 580.432,
      "eval_steps_per_second": 12.17,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 98172.5625,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.5137,
      "eval_samples_per_second": 562.235,
      "eval_steps_per_second": 11.789,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.3552,
      "eval_samples_per_second": 578.874,
      "eval_steps_per_second": 12.138,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 512333.03125,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.7027,
      "eval_samples_per_second": 543.603,
      "eval_steps_per_second": 11.398,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 1070884.875,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.4103,
      "eval_samples_per_second": 572.98,
      "eval_steps_per_second": 12.014,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.2968,
      "eval_samples_per_second": 585.264,
      "eval_steps_per_second": 12.272,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 1487442.5,
      "learning_rate": 6.02375960866527e-06,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.6095,
      "eval_samples_per_second": 552.638,
      "eval_steps_per_second": 11.588,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 1775379.0,
      "learning_rate": 2.5296995108315863e-06,
      "loss": 0.0,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7080592966217989,
    "num_train_epochs": 9,
    "temperature": 10
  }
}