trainer_state.json

{
  "best_metric": 0.6176406741142273,
  "best_model_checkpoint": "hBERTv1_new_pretrain_cola/checkpoint-201",
  "epoch": 8.0,
  "global_step": 536,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6331,
      "step": 67
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.618148148059845,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8818,
      "eval_samples_per_second": 554.271,
      "eval_steps_per_second": 4.783,
      "step": 67
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.624,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.620297908782959,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8844,
      "eval_samples_per_second": 553.481,
      "eval_steps_per_second": 4.776,
      "step": 134
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6173,
      "step": 201
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6176406741142273,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8882,
      "eval_samples_per_second": 552.379,
      "eval_steps_per_second": 4.766,
      "step": 201
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6176,
      "step": 268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.618543267250061,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8866,
      "eval_samples_per_second": 552.849,
      "eval_steps_per_second": 4.771,
      "step": 268
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6121,
      "step": 335
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6193671822547913,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8856,
      "eval_samples_per_second": 553.129,
      "eval_steps_per_second": 4.773,
      "step": 335
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.6112,
      "step": 402
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6186181902885437,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.885,
      "eval_samples_per_second": 553.315,
      "eval_steps_per_second": 4.775,
      "step": 402
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.6132,
      "step": 469
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.626687228679657,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8814,
      "eval_samples_per_second": 554.386,
      "eval_steps_per_second": 4.784,
      "step": 469
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.6124,
      "step": 536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6217716336250305,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8821,
      "eval_samples_per_second": 554.165,
      "eval_steps_per_second": 4.782,
      "step": 536
    },
    {
      "epoch": 8.0,
      "step": 536,
      "total_flos": 1.0131004291284992e+16,
      "train_loss": 0.6175890894078496,
      "train_runtime": 442.252,
      "train_samples_per_second": 966.756,
      "train_steps_per_second": 7.575
    }
  ],
  "max_steps": 3350,
  "num_train_epochs": 50,
  "total_flos": 1.0131004291284992e+16,
  "trial_name": null,
  "trial_params": null
}