trainer_state.json

{
  "best_metric": 0.012019365094602108,
  "best_model_checkpoint": "/projects/users/mgillele/trash/test_segmenter/2025-10-13T18:02:46.689601_BERT_model_paper/models/checkpoint-498",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1494,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.08617644011974335,
      "learning_rate": 4.9e-05,
      "loss": 0.0243,
      "step": 498
    },
    {
      "epoch": 1.0,
      "eval_accuracy": {
        "accuracy": 0.9797698824124416
      },
      "eval_f1": [
        0.9887650969010392,
        0.8985200845665962
      ],
      "eval_loss": 0.012019365094602108,
      "eval_precision": [
        0.9864561927890902,
        0.91792656587473
      ],
      "eval_recall": [
        0.9910848348348348,
        0.8799171842650103
      ],
      "eval_runtime": 4.235,
      "eval_samples_per_second": 234.946,
      "eval_steps_per_second": 7.556,
      "step": 498
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.09689918160438538,
      "learning_rate": 4.8e-05,
      "loss": 0.0094,
      "step": 996
    },
    {
      "epoch": 2.0,
      "eval_accuracy": {
        "accuracy": 0.9800368075750552
      },
      "eval_f1": [
        0.9889099607439146,
        0.9001335301145548
      ],
      "eval_loss": 0.012484115548431873,
      "eval_precision": [
        0.9868996993628986,
        0.9169530355097365
      ],
      "eval_recall": [
        0.9909284284284284,
        0.88391994478951
      ],
      "eval_runtime": 4.179,
      "eval_samples_per_second": 238.098,
      "eval_steps_per_second": 7.657,
      "step": 996
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.2906811535358429,
      "learning_rate": 4.7e-05,
      "loss": 0.0064,
      "step": 1494
    },
    {
      "epoch": 3.0,
      "eval_accuracy": {
        "accuracy": 0.979011253002908
      },
      "eval_f1": [
        0.9882948384468332,
        0.8986078098471987,
        0.0
      ],
      "eval_loss": 0.014125695452094078,
      "eval_precision": [
        0.9901255886970173,
        0.8844919786096257,
        0.0
      ],
      "eval_recall": [
        0.9864708458458459,
        0.9131815044858523,
        0.0
      ],
      "eval_runtime": 4.1316,
      "eval_samples_per_second": 240.828,
      "eval_steps_per_second": 7.745,
      "step": 1494
    }
  ],
  "logging_steps": 500,
  "max_steps": 24900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 10,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8436689183872296.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}