trainer_state.json

{
  "best_metric": 0.6188952922821045,
  "best_model_checkpoint": "bert_base_lda_20_v1_cola/checkpoint-68",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 238,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4452758729457855,
      "learning_rate": 0.00098,
      "loss": 0.7571,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6196237802505493,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6803,
      "eval_samples_per_second": 1533.141,
      "eval_steps_per_second": 7.35,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9407540559768677,
      "learning_rate": 0.00096,
      "loss": 0.613,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6188952922821045,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6631,
      "eval_samples_per_second": 1572.981,
      "eval_steps_per_second": 7.541,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0614078044891357,
      "learning_rate": 0.00094,
      "loss": 0.6097,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6194814443588257,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6519,
      "eval_samples_per_second": 1599.847,
      "eval_steps_per_second": 7.669,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6363035440444946,
      "learning_rate": 0.00092,
      "loss": 0.6106,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6196293830871582,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6778,
      "eval_samples_per_second": 1538.719,
      "eval_steps_per_second": 7.376,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9160279035568237,
      "learning_rate": 0.0009000000000000001,
      "loss": 0.6114,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6194814443588257,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6776,
      "eval_samples_per_second": 1539.156,
      "eval_steps_per_second": 7.379,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.48956847190856934,
      "learning_rate": 0.00088,
      "loss": 0.6094,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6189964413642883,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6726,
      "eval_samples_per_second": 1550.586,
      "eval_steps_per_second": 7.433,
      "step": 204
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.5084835290908813,
      "learning_rate": 0.00086,
      "loss": 0.6102,
      "step": 238
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.619966447353363,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6632,
      "eval_samples_per_second": 1572.699,
      "eval_steps_per_second": 7.539,
      "step": 238
    },
    {
      "epoch": 7.0,
      "step": 238,
      "total_flos": 7874519220341760.0,
      "train_loss": 0.6316283650758887,
      "train_runtime": 116.2402,
      "train_samples_per_second": 3678.16,
      "train_steps_per_second": 14.625
    }
  ],
  "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": 7874519220341760.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}