trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6166294813156128,
  "best_model_checkpoint": "tiny_bert_rand_100_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": 0.732401430606842,
      "learning_rate": 4.9e-05,
      "loss": 0.6142,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6166294813156128,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3149,
      "eval_samples_per_second": 3312.123,
      "eval_steps_per_second": 15.878,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8920761942863464,
      "learning_rate": 4.8e-05,
      "loss": 0.6003,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6185376644134521,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3515,
      "eval_samples_per_second": 2967.071,
      "eval_steps_per_second": 14.224,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.510546088218689,
      "learning_rate": 4.7e-05,
      "loss": 0.5794,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6874400973320007,
      "eval_loss": 0.621528685092926,
      "eval_matthews_correlation": 0.017906650023835943,
      "eval_runtime": 0.3282,
      "eval_samples_per_second": 3178.369,
      "eval_steps_per_second": 15.237,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0394983291625977,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5396,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6711409687995911,
      "eval_loss": 0.6723660826683044,
      "eval_matthews_correlation": 0.059735727206595625,
      "eval_runtime": 0.3379,
      "eval_samples_per_second": 3086.867,
      "eval_steps_per_second": 14.798,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.666203260421753,
      "learning_rate": 4.5e-05,
      "loss": 0.4923,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6452540755271912,
      "eval_loss": 0.6689262390136719,
      "eval_matthews_correlation": 0.07858985543075096,
      "eval_runtime": 0.3178,
      "eval_samples_per_second": 3281.625,
      "eval_steps_per_second": 15.732,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.4024789333343506,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4541,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6260786056518555,
      "eval_loss": 0.7376295924186707,
      "eval_matthews_correlation": 0.08036910527448742,
      "eval_runtime": 0.3091,
      "eval_samples_per_second": 3374.438,
      "eval_steps_per_second": 16.177,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 1345426898614272.0,
      "train_loss": 0.5466599370919022,
      "train_runtime": 34.0274,
      "train_samples_per_second": 12564.883,
      "train_steps_per_second": 49.96
    }
  ],
  "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": 1345426898614272.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}