trainer_state.json

{
  "best_metric": 0.9150529503822327,
  "best_model_checkpoint": "hBERTv2_new_pretrain_wnli/checkpoint-10",
  "epoch": 7.0,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.00049,
      "loss": 8.8479,
      "step": 5
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 1.9492391347885132,
      "eval_runtime": 0.1327,
      "eval_samples_per_second": 535.165,
      "eval_steps_per_second": 7.538,
      "step": 5
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048,
      "loss": 9.5465,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.9150529503822327,
      "eval_runtime": 0.1358,
      "eval_samples_per_second": 522.721,
      "eval_steps_per_second": 7.362,
      "step": 10
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047,
      "loss": 8.6319,
      "step": 15
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 2.451029062271118,
      "eval_runtime": 0.1296,
      "eval_samples_per_second": 547.845,
      "eval_steps_per_second": 7.716,
      "step": 15
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046,
      "loss": 9.7722,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 4.1092000007629395,
      "eval_runtime": 0.1294,
      "eval_samples_per_second": 548.875,
      "eval_steps_per_second": 7.731,
      "step": 20
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045000000000000004,
      "loss": 10.479,
      "step": 25
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 4.987680912017822,
      "eval_runtime": 0.1316,
      "eval_samples_per_second": 539.47,
      "eval_steps_per_second": 7.598,
      "step": 25
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044,
      "loss": 10.4548,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 5.494718551635742,
      "eval_runtime": 0.1299,
      "eval_samples_per_second": 546.636,
      "eval_steps_per_second": 7.699,
      "step": 30
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.00043,
      "loss": 10.5408,
      "step": 35
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 4.8345136642456055,
      "eval_runtime": 0.136,
      "eval_samples_per_second": 521.971,
      "eval_steps_per_second": 7.352,
      "step": 35
    },
    {
      "epoch": 7.0,
      "step": 35,
      "total_flos": 650477628817408.0,
      "train_loss": 9.753294808523995,
      "train_runtime": 48.7913,
      "train_samples_per_second": 650.731,
      "train_steps_per_second": 5.124
    }
  ],
  "max_steps": 250,
  "num_train_epochs": 50,
  "total_flos": 650477628817408.0,
  "trial_name": null,
  "trial_params": null
}