trainer_state.json

{
  "best_metric": 0.6307324171066284,
  "best_model_checkpoint": "hBERTv1_new_pretrain_48_emb_com_cola/checkpoint-134",
  "epoch": 7.0,
  "global_step": 469,
  "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": 0.639,
      "step": 67
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6325676441192627,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 4.7202,
      "eval_samples_per_second": 220.963,
      "eval_steps_per_second": 1.907,
      "step": 67
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.608,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6307324171066284,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 4.715,
      "eval_samples_per_second": 221.209,
      "eval_steps_per_second": 1.909,
      "step": 134
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6029,
      "step": 201
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6311112642288208,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 4.707,
      "eval_samples_per_second": 221.585,
      "eval_steps_per_second": 1.912,
      "step": 201
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.5893,
      "step": 268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6836050152778625,
      "eval_loss": 0.6356986165046692,
      "eval_matthews_correlation": 0.00300496871017747,
      "eval_runtime": 4.7054,
      "eval_samples_per_second": 221.658,
      "eval_steps_per_second": 1.913,
      "step": 268
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.5692,
      "step": 335
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6605944633483887,
      "eval_loss": 0.6723994612693787,
      "eval_matthews_correlation": 0.026963793941824566,
      "eval_runtime": 4.7007,
      "eval_samples_per_second": 221.882,
      "eval_steps_per_second": 1.915,
      "step": 335
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.5476,
      "step": 402
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6548417806625366,
      "eval_loss": 0.6516734957695007,
      "eval_matthews_correlation": 0.08300801435614485,
      "eval_runtime": 4.7009,
      "eval_samples_per_second": 221.87,
      "eval_steps_per_second": 1.915,
      "step": 402
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.5294,
      "step": 469
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6855225563049316,
      "eval_loss": 0.6674203276634216,
      "eval_matthews_correlation": 0.06864834703912359,
      "eval_runtime": 4.7015,
      "eval_samples_per_second": 221.845,
      "eval_steps_per_second": 1.914,
      "step": 469
    },
    {
      "epoch": 7.0,
      "step": 469,
      "total_flos": 8873620000473088.0,
      "train_loss": 0.5836383193286497,
      "train_runtime": 969.9748,
      "train_samples_per_second": 440.785,
      "train_steps_per_second": 3.454
    }
  ],
  "max_steps": 3350,
  "num_train_epochs": 50,
  "total_flos": 8873620000473088.0,
  "trial_name": null,
  "trial_params": null
}