{
  "best_metric": 0.20824170112609863,
  "best_model_checkpoint": "distilbert-base-uncased-imdb/checkpoint-1250",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "grad_norm": 3.633013963699341,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2727,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9178,
      "eval_f1": 0.9211283822682786,
      "eval_loss": 0.2193804234266281,
      "eval_precision": 0.8852821836960532,
      "eval_recall": 0.96,
      "eval_runtime": 86.6891,
      "eval_samples_per_second": 57.677,
      "eval_steps_per_second": 1.811,
      "step": 625
    },
    {
      "epoch": 1.6,
      "grad_norm": 16.50516700744629,
      "learning_rate": 1.2e-05,
      "loss": 0.1569,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9236,
      "eval_f1": 0.9261978361669243,
      "eval_loss": 0.20824170112609863,
      "eval_precision": 0.8957399103139013,
      "eval_recall": 0.9588,
      "eval_runtime": 86.8325,
      "eval_samples_per_second": 57.582,
      "eval_steps_per_second": 1.808,
      "step": 1250
    },
    {
      "epoch": 2.4,
      "grad_norm": 13.776262283325195,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1211,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9256,
      "eval_f1": 0.9245436105476673,
      "eval_loss": 0.2738637328147888,
      "eval_precision": 0.9378600823045268,
      "eval_recall": 0.9116,
      "eval_runtime": 86.4733,
      "eval_samples_per_second": 57.821,
      "eval_steps_per_second": 1.816,
      "step": 1875
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.0917992889881134,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0753,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.31518808007240295,
      "learning_rate": 0.0,
      "loss": 0.0544,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9284,
      "eval_f1": 0.9284286285485805,
      "eval_loss": 0.28977659344673157,
      "eval_precision": 0.9280575539568345,
      "eval_recall": 0.9288,
      "eval_runtime": 84.4716,
      "eval_samples_per_second": 59.191,
      "eval_steps_per_second": 1.859,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "step": 2500,
      "total_flos": 1.059739189248e+16,
      "train_loss": 0.13609123306274415,
      "train_runtime": 4154.1938,
      "train_samples_per_second": 19.258,
      "train_steps_per_second": 0.602
    }
  ],
  "logging_steps": 500,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 1.059739189248e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}