trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6219514012336731,
  "best_model_checkpoint": "tiny_bert_km_50_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": 0.6227364540100098,
      "learning_rate": 4.9e-05,
      "loss": 0.6193,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6219514012336731,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3098,
      "eval_samples_per_second": 3366.347,
      "eval_steps_per_second": 16.138,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0430492162704468,
      "learning_rate": 4.8e-05,
      "loss": 0.6051,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6245852112770081,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3171,
      "eval_samples_per_second": 3289.089,
      "eval_steps_per_second": 15.767,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1225727796554565,
      "learning_rate": 4.7e-05,
      "loss": 0.5944,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6285073757171631,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3196,
      "eval_samples_per_second": 3263.762,
      "eval_steps_per_second": 15.646,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.3095027208328247,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5715,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6778523325920105,
      "eval_loss": 0.6386438012123108,
      "eval_matthews_correlation": 0.05783382654286213,
      "eval_runtime": 0.3201,
      "eval_samples_per_second": 3257.987,
      "eval_steps_per_second": 15.618,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.773003101348877,
      "learning_rate": 4.5e-05,
      "loss": 0.531,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6586769223213196,
      "eval_loss": 0.6561414003372192,
      "eval_matthews_correlation": 0.03871993486842746,
      "eval_runtime": 0.3232,
      "eval_samples_per_second": 3227.364,
      "eval_steps_per_second": 15.472,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.923257827758789,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4788,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6529242396354675,
      "eval_loss": 0.7090504169464111,
      "eval_matthews_correlation": 0.06667066349742735,
      "eval_runtime": 0.3639,
      "eval_samples_per_second": 2866.541,
      "eval_steps_per_second": 13.742,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 1345426898614272.0,
      "train_loss": 0.5666545886619418,
      "train_runtime": 33.9821,
      "train_samples_per_second": 12581.606,
      "train_steps_per_second": 50.026
    }
  ],
  "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
}