trainer_state.json

{
  "best_metric": 1.0963075160980225,
  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mnli/checkpoint-6136",
  "epoch": 7.0,
  "global_step": 21476,
  "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": 1.1346,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0987932682037354,
      "eval_runtime": 17.7369,
      "eval_samples_per_second": 553.367,
      "eval_steps_per_second": 4.341,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048,
      "loss": 1.1013,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0963075160980225,
      "eval_runtime": 17.7358,
      "eval_samples_per_second": 553.402,
      "eval_steps_per_second": 4.342,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047,
      "loss": 1.1611,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0990049839019775,
      "eval_runtime": 17.7014,
      "eval_samples_per_second": 554.477,
      "eval_steps_per_second": 4.35,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046,
      "loss": 1.0988,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0987932682037354,
      "eval_runtime": 17.7275,
      "eval_samples_per_second": 553.658,
      "eval_steps_per_second": 4.344,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045000000000000004,
      "loss": 1.0987,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0987932682037354,
      "eval_runtime": 17.7181,
      "eval_samples_per_second": 553.953,
      "eval_steps_per_second": 4.346,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044,
      "loss": 1.0985,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0990049839019775,
      "eval_runtime": 17.7077,
      "eval_samples_per_second": 554.278,
      "eval_steps_per_second": 4.348,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.00043,
      "loss": 1.0987,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0990049839019775,
      "eval_runtime": 17.6862,
      "eval_samples_per_second": 554.953,
      "eval_steps_per_second": 4.354,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "step": 21476,
      "total_flos": 4.071085382597345e+17,
      "train_loss": 1.1130940406587586,
      "train_runtime": 14060.0313,
      "train_samples_per_second": 1396.519,
      "train_steps_per_second": 10.91
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 4.071085382597345e+17,
  "trial_name": null,
  "trial_params": null
}