trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6197782754898071,
  "best_model_checkpoint": "bert_base_rand_10_v2_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.760583758354187,
      "learning_rate": 4.9e-05,
      "loss": 0.6124,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6197782754898071,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 1535.653,
      "eval_steps_per_second": 7.362,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5169968605041504,
      "learning_rate": 4.8e-05,
      "loss": 0.5889,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.690316379070282,
      "eval_loss": 0.6246058344841003,
      "eval_matthews_correlation": 0.04129856774209186,
      "eval_runtime": 0.699,
      "eval_samples_per_second": 1492.229,
      "eval_steps_per_second": 7.154,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9388715028762817,
      "learning_rate": 4.7e-05,
      "loss": 0.5412,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6845637559890747,
      "eval_loss": 0.6387814283370972,
      "eval_matthews_correlation": 0.029692033315243296,
      "eval_runtime": 0.674,
      "eval_samples_per_second": 1547.589,
      "eval_steps_per_second": 7.419,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.953024864196777,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4976,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6558005809783936,
      "eval_loss": 0.7266033887863159,
      "eval_matthews_correlation": 0.12227142135867913,
      "eval_runtime": 0.6895,
      "eval_samples_per_second": 1512.593,
      "eval_steps_per_second": 7.251,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.9303916692733765,
      "learning_rate": 4.5e-05,
      "loss": 0.4461,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6730585098266602,
      "eval_loss": 0.6882631778717041,
      "eval_matthews_correlation": 0.12232784170821964,
      "eval_runtime": 0.6714,
      "eval_samples_per_second": 1553.494,
      "eval_steps_per_second": 7.447,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.0536789894104,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4039,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6433365345001221,
      "eval_loss": 0.8319432139396667,
      "eval_matthews_correlation": 0.078501645978455,
      "eval_runtime": 0.6953,
      "eval_samples_per_second": 1500.174,
      "eval_steps_per_second": 7.192,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5150212961084703,
      "train_runtime": 100.0509,
      "train_samples_per_second": 4273.326,
      "train_steps_per_second": 16.991
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6749587903150080.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}