trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6205900311470032,
  "best_model_checkpoint": "bert_base_rand_5_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": 0.6443101167678833,
      "learning_rate": 4.9e-05,
      "loss": 0.6133,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6205900311470032,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6841,
      "eval_samples_per_second": 1524.567,
      "eval_steps_per_second": 7.309,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.772774577140808,
      "learning_rate": 4.8e-05,
      "loss": 0.5905,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.6222543120384216,
      "eval_matthews_correlation": 0.03479330099568123,
      "eval_runtime": 0.7321,
      "eval_samples_per_second": 1424.746,
      "eval_steps_per_second": 6.83,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9345377683639526,
      "learning_rate": 4.7e-05,
      "loss": 0.539,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6625120043754578,
      "eval_loss": 0.6405149698257446,
      "eval_matthews_correlation": 0.09673744644490058,
      "eval_runtime": 0.6726,
      "eval_samples_per_second": 1550.663,
      "eval_steps_per_second": 7.434,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.012849807739258,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4892,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.641418993473053,
      "eval_loss": 0.6828953623771667,
      "eval_matthews_correlation": 0.11692964845499106,
      "eval_runtime": 0.6962,
      "eval_samples_per_second": 1498.026,
      "eval_steps_per_second": 7.181,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.711766004562378,
      "learning_rate": 4.5e-05,
      "loss": 0.4391,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6701821684837341,
      "eval_loss": 0.677309513092041,
      "eval_matthews_correlation": 0.09430442091064549,
      "eval_runtime": 0.678,
      "eval_samples_per_second": 1538.363,
      "eval_steps_per_second": 7.375,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.584491014480591,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3908,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6500479578971863,
      "eval_loss": 0.8017939329147339,
      "eval_matthews_correlation": 0.0495086786186171,
      "eval_runtime": 0.6907,
      "eval_samples_per_second": 1510.048,
      "eval_steps_per_second": 7.239,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5102848679411645,
      "train_runtime": 99.5545,
      "train_samples_per_second": 4294.633,
      "train_steps_per_second": 17.076
    }
  ],
  "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
}