trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6176116466522217,
  "best_model_checkpoint": "bert_base_rand_100_v1_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": 1.2854377031326294,
      "learning_rate": 4.9e-05,
      "loss": 0.6133,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6176116466522217,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6775,
      "eval_samples_per_second": 1539.593,
      "eval_steps_per_second": 7.381,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.719799518585205,
      "learning_rate": 4.8e-05,
      "loss": 0.5913,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6883988380432129,
      "eval_loss": 0.6251381039619446,
      "eval_matthews_correlation": -0.03589254563226399,
      "eval_runtime": 0.6763,
      "eval_samples_per_second": 1542.319,
      "eval_steps_per_second": 7.394,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8317792415618896,
      "learning_rate": 4.7e-05,
      "loss": 0.5374,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6481304168701172,
      "eval_loss": 0.6604347825050354,
      "eval_matthews_correlation": 0.10041605806534921,
      "eval_runtime": 0.684,
      "eval_samples_per_second": 1524.797,
      "eval_steps_per_second": 7.31,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.593310356140137,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.486,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6567593216896057,
      "eval_loss": 0.7520363330841064,
      "eval_matthews_correlation": 0.08852460407026867,
      "eval_runtime": 0.7103,
      "eval_samples_per_second": 1468.365,
      "eval_steps_per_second": 7.039,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2590205669403076,
      "learning_rate": 4.5e-05,
      "loss": 0.4227,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.7340983748435974,
      "eval_matthews_correlation": 0.13750626640004746,
      "eval_runtime": 0.684,
      "eval_samples_per_second": 1524.759,
      "eval_steps_per_second": 7.309,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.552171468734741,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3742,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6567593216896057,
      "eval_loss": 0.7870461940765381,
      "eval_matthews_correlation": 0.11941328770402174,
      "eval_runtime": 0.6662,
      "eval_samples_per_second": 1565.555,
      "eval_steps_per_second": 7.505,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5041563651140999,
      "train_runtime": 102.8186,
      "train_samples_per_second": 4158.296,
      "train_steps_per_second": 16.534
    }
  ],
  "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
}