checkpoint-122/trainer_state.json

{
  "best_global_step": 122,
  "best_metric": 0.9440661564177004,
  "best_model_checkpoint": "../models/distilbert-ad-detection/checkpoint-122",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 122,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.819672131147541,
      "grad_norm": 1.6779587268829346,
      "learning_rate": 2.45e-05,
      "loss": 0.5205,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8518518518518519,
      "eval_f1": 0.7837037037037037,
      "eval_loss": 0.28867536783218384,
      "eval_precision": 0.7256515775034293,
      "eval_recall": 0.8518518518518519,
      "eval_runtime": 1.7374,
      "eval_samples_per_second": 139.867,
      "eval_steps_per_second": 9.209,
      "step": 61
    },
    {
      "epoch": 1.639344262295082,
      "grad_norm": 1.3054509162902832,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 0.229,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9465020576131687,
      "eval_f1": 0.9440661564177004,
      "eval_loss": 0.14445427060127258,
      "eval_precision": 0.9448681682004273,
      "eval_recall": 0.9465020576131687,
      "eval_runtime": 1.1734,
      "eval_samples_per_second": 207.093,
      "eval_steps_per_second": 13.636,
      "step": 122
    }
  ],
  "logging_steps": 50,
  "max_steps": 183,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 26602144394160.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}