trainer_state.json

{
  "best_global_step": 3122,
  "best_metric": 0.5257779194390215,
  "best_model_checkpoint": "outputs/bilstm-spam-binary/checkpoint-3122",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 3122,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4966887831687927,
      "learning_rate": 0.0009997543852189499,
      "loss": 0.5757,
      "step": 446
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7290485192186515,
      "eval_f1": 0.42164723032069973,
      "eval_loss": 0.5851318836212158,
      "eval_precision": 0.3645242596093258,
      "eval_recall": 0.5,
      "eval_runtime": 6.2523,
      "eval_samples_per_second": 253.825,
      "eval_steps_per_second": 7.997,
      "step": 446
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.35379767417907715,
      "learning_rate": 0.0009990155744361903,
      "loss": 0.5739,
      "step": 892
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7290485192186515,
      "eval_f1": 0.42164723032069973,
      "eval_loss": 0.5852128267288208,
      "eval_precision": 0.3645242596093258,
      "eval_recall": 0.5,
      "eval_runtime": 6.2531,
      "eval_samples_per_second": 253.796,
      "eval_steps_per_second": 7.996,
      "step": 892
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5833616256713867,
      "learning_rate": 0.0009977842955283425,
      "loss": 0.5726,
      "step": 1338
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7290485192186515,
      "eval_f1": 0.42164723032069973,
      "eval_loss": 0.5832122564315796,
      "eval_precision": 0.3645242596093258,
      "eval_recall": 0.5,
      "eval_runtime": 6.2546,
      "eval_samples_per_second": 253.734,
      "eval_steps_per_second": 7.994,
      "step": 1338
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9242071509361267,
      "learning_rate": 0.0009960617636190344,
      "loss": 0.5606,
      "step": 1784
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7290485192186515,
      "eval_f1": 0.42164723032069973,
      "eval_loss": 0.5644996762275696,
      "eval_precision": 0.3645242596093258,
      "eval_recall": 0.5,
      "eval_runtime": 6.2529,
      "eval_samples_per_second": 253.801,
      "eval_steps_per_second": 7.996,
      "step": 1784
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6944896578788757,
      "learning_rate": 0.0009938496786392973,
      "loss": 0.5453,
      "step": 2230
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7321991178323881,
      "eval_f1": 0.4810744810744811,
      "eval_loss": 0.563392698764801,
      "eval_precision": 0.6428808639393865,
      "eval_recall": 0.5233492794114691,
      "eval_runtime": 6.2646,
      "eval_samples_per_second": 253.33,
      "eval_steps_per_second": 7.981,
      "step": 2230
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.42335620522499084,
      "learning_rate": 0.0009911502236499372,
      "loss": 0.5441,
      "step": 2676
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7284183994959043,
      "eval_f1": 0.4236885275598579,
      "eval_loss": 0.5696581602096558,
      "eval_precision": 0.53125,
      "eval_recall": 0.5002984864625837,
      "eval_runtime": 6.2683,
      "eval_samples_per_second": 253.179,
      "eval_steps_per_second": 7.977,
      "step": 2676
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.6146748661994934,
      "learning_rate": 0.0009879660626871199,
      "loss": 0.5431,
      "step": 3122
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7353497164461248,
      "eval_f1": 0.5257779194390215,
      "eval_loss": 0.5626002550125122,
      "eval_precision": 0.6485838779956428,
      "eval_recall": 0.5452372816626802,
      "eval_runtime": 6.2648,
      "eval_samples_per_second": 253.32,
      "eval_steps_per_second": 7.981,
      "step": 3122
    }
  ],
  "logging_steps": 446,
  "max_steps": 44600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "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": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}