{
  "best_global_step": 8236,
  "best_metric": 0.8473409433764713,
  "best_model_checkpoint": "./camelbert-ner-author/checkpoint-8236",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 8236,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9559553861618042,
      "learning_rate": 1.5002428363283148e-05,
      "loss": 0.6744,
      "step": 2059
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.830406543180178,
      "eval_f1_macro": 0.791267966245864,
      "eval_f1_micro": 0.830406543180178,
      "eval_loss": 0.5986002683639526,
      "eval_precision_macro": 0.7902411997408786,
      "eval_precision_micro": 0.830406543180178,
      "eval_recall_macro": 0.8145205534902739,
      "eval_recall_micro": 0.830406543180178,
      "eval_runtime": 75.6826,
      "eval_samples_per_second": 54.927,
      "eval_steps_per_second": 3.435,
      "step": 2059
    },
    {
      "epoch": 2.0,
      "grad_norm": 60.31794357299805,
      "learning_rate": 1.0002428363283147e-05,
      "loss": 0.1155,
      "step": 4118
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8448400288669714,
      "eval_f1_macro": 0.7972921054481455,
      "eval_f1_micro": 0.8448400288669714,
      "eval_loss": 0.7185700535774231,
      "eval_precision_macro": 0.8081525378723841,
      "eval_precision_micro": 0.8448400288669714,
      "eval_recall_macro": 0.8163042932269564,
      "eval_recall_micro": 0.8448400288669714,
      "eval_runtime": 75.1555,
      "eval_samples_per_second": 55.312,
      "eval_steps_per_second": 3.459,
      "step": 4118
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.02701091766357422,
      "learning_rate": 5.002428363283147e-06,
      "loss": 0.0355,
      "step": 6177
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8857349049795525,
      "eval_f1_macro": 0.8350373600980577,
      "eval_f1_micro": 0.8857349049795525,
      "eval_loss": 0.5690950155258179,
      "eval_precision_macro": 0.8303194562290555,
      "eval_precision_micro": 0.8857349049795525,
      "eval_recall_macro": 0.8522223907532204,
      "eval_recall_micro": 0.8857349049795525,
      "eval_runtime": 75.1931,
      "eval_samples_per_second": 55.284,
      "eval_steps_per_second": 3.458,
      "step": 6177
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.015318239107728004,
      "learning_rate": 2.428363283147159e-09,
      "loss": 0.0091,
      "step": 8236
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8951166706759682,
      "eval_f1_macro": 0.8473409433764713,
      "eval_f1_micro": 0.8951166706759682,
      "eval_loss": 0.5703061819076538,
      "eval_precision_macro": 0.8399460951777158,
      "eval_precision_micro": 0.8951166706759682,
      "eval_recall_macro": 0.8710852747128797,
      "eval_recall_micro": 0.8951166706759682,
      "eval_runtime": 75.5896,
      "eval_samples_per_second": 54.994,
      "eval_steps_per_second": 3.44,
      "step": 8236
    }
  ],
  "logging_steps": 500,
  "max_steps": 8236,
  "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": 3.467447932454093e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}