trainer_state.json

{
  "best_global_step": 100,
  "best_metric": 0.9942340850830078,
  "best_model_checkpoint": "./output_deep_fast\\checkpoint-100",
  "epoch": 0.3551609322974473,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07103218645948946,
      "grad_norm": 0.5432878136634827,
      "learning_rate": 4.470588235294118e-05,
      "loss": 1.2951,
      "step": 20
    },
    {
      "epoch": 0.14206437291897892,
      "grad_norm": 0.40483182668685913,
      "learning_rate": 9.176470588235295e-05,
      "loss": 1.0208,
      "step": 40
    },
    {
      "epoch": 0.21309655937846836,
      "grad_norm": 0.5101248025894165,
      "learning_rate": 0.00013882352941176472,
      "loss": 0.9941,
      "step": 60
    },
    {
      "epoch": 0.28412874583795783,
      "grad_norm": 0.40778905153274536,
      "learning_rate": 0.00018588235294117648,
      "loss": 0.8642,
      "step": 80
    },
    {
      "epoch": 0.3551609322974473,
      "grad_norm": 0.3639456629753113,
      "learning_rate": 0.00019983303108908946,
      "loss": 0.8981,
      "step": 100
    },
    {
      "epoch": 0.3551609322974473,
      "eval_loss": 0.9942340850830078,
      "eval_runtime": 100.6429,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 4.769,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 846,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8014109808190464.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}