{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2544,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5738709677419355,
      "eval_loss": 0.1914977878332138,
      "eval_runtime": 5.3865,
      "eval_samples_per_second": 575.517,
      "eval_steps_per_second": 12.067,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5165097713470459,
      "learning_rate": 1.606918238993711e-05,
      "loss": 0.3072,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8141935483870968,
      "eval_loss": 0.09490247070789337,
      "eval_runtime": 5.6092,
      "eval_samples_per_second": 552.66,
      "eval_steps_per_second": 11.588,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8741935483870967,
      "eval_loss": 0.06484736502170563,
      "eval_runtime": 6.0767,
      "eval_samples_per_second": 510.146,
      "eval_steps_per_second": 10.697,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.43419399857521057,
      "learning_rate": 1.2138364779874214e-05,
      "loss": 0.1097,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8958064516129032,
      "eval_loss": 0.05110108479857445,
      "eval_runtime": 5.4249,
      "eval_samples_per_second": 571.437,
      "eval_steps_per_second": 11.982,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.2985864579677582,
      "learning_rate": 8.207547169811321e-06,
      "loss": 0.072,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9048387096774193,
      "eval_loss": 0.04280891269445419,
      "eval_runtime": 5.5512,
      "eval_samples_per_second": 558.437,
      "eval_steps_per_second": 11.709,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9074193548387097,
      "eval_loss": 0.038190603256225586,
      "eval_runtime": 5.4902,
      "eval_samples_per_second": 564.646,
      "eval_steps_per_second": 11.839,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.2691451609134674,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0579,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9116129032258065,
      "eval_loss": 0.03593328222632408,
      "eval_runtime": 5.5022,
      "eval_samples_per_second": 563.415,
      "eval_steps_per_second": 11.814,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.265292763710022,
      "learning_rate": 3.459119496855346e-07,
      "loss": 0.0522,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2544,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5962321311067933,
    "num_train_epochs": 8,
    "temperature": 20
  }
}