{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.433962264150944,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6309677419354839,
      "eval_loss": 0.2469712644815445,
      "eval_runtime": 5.4589,
      "eval_samples_per_second": 567.879,
      "eval_steps_per_second": 11.907,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5992535352706909,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.3971,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8422580645161291,
      "eval_loss": 0.10715335607528687,
      "eval_runtime": 5.8957,
      "eval_samples_per_second": 525.809,
      "eval_steps_per_second": 11.025,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8893548387096775,
      "eval_loss": 0.06534729152917862,
      "eval_runtime": 5.5645,
      "eval_samples_per_second": 557.106,
      "eval_steps_per_second": 11.681,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.5163053870201111,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1257,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9077419354838709,
      "eval_loss": 0.04786108061671257,
      "eval_runtime": 5.7341,
      "eval_samples_per_second": 540.622,
      "eval_steps_per_second": 11.336,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.3360358476638794,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0738,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9161290322580645,
      "eval_loss": 0.038218673318624496,
      "eval_runtime": 5.7603,
      "eval_samples_per_second": 538.164,
      "eval_steps_per_second": 11.284,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9258064516129032,
      "eval_loss": 0.03327132761478424,
      "eval_runtime": 5.565,
      "eval_samples_per_second": 557.05,
      "eval_steps_per_second": 11.68,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.2815706431865692,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.056,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9296774193548387,
      "eval_loss": 0.03050699457526207,
      "eval_runtime": 5.8127,
      "eval_samples_per_second": 533.313,
      "eval_steps_per_second": 11.182,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.3059941530227661,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0481,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9303225806451613,
      "eval_loss": 0.028380775824189186,
      "eval_runtime": 5.5569,
      "eval_samples_per_second": 557.861,
      "eval_steps_per_second": 11.697,
      "step": 2544
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.932258064516129,
      "eval_loss": 0.02733077108860016,
      "eval_runtime": 5.4623,
      "eval_samples_per_second": 567.527,
      "eval_steps_per_second": 11.9,
      "step": 2862
    },
    {
      "epoch": 9.433962264150944,
      "grad_norm": 0.27681389451026917,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0446,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 780738843279612.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.1299955797744864,
    "num_train_epochs": 10,
    "temperature": 4
  }
}