trainer_state.json

{
  "best_metric": 0.6930931806564331,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_qnli_256/checkpoint-2457",
  "epoch": 8.0,
  "global_step": 6552,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.6933,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6931739449501038,
      "eval_runtime": 9.707,
      "eval_samples_per_second": 562.788,
      "eval_steps_per_second": 4.43,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.6932,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6932452321052551,
      "eval_runtime": 9.6752,
      "eval_samples_per_second": 564.64,
      "eval_steps_per_second": 4.444,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.6932,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930931806564331,
      "eval_runtime": 9.8332,
      "eval_samples_per_second": 555.564,
      "eval_steps_per_second": 4.373,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6932,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6932864189147949,
      "eval_runtime": 9.8612,
      "eval_samples_per_second": 553.991,
      "eval_steps_per_second": 4.361,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.6932,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931220889091492,
      "eval_runtime": 9.7636,
      "eval_samples_per_second": 559.525,
      "eval_steps_per_second": 4.404,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6932,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931174993515015,
      "eval_runtime": 9.7449,
      "eval_samples_per_second": 560.6,
      "eval_steps_per_second": 4.413,
      "step": 4914
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.6932,
      "step": 5733
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930953860282898,
      "eval_runtime": 9.7483,
      "eval_samples_per_second": 560.404,
      "eval_steps_per_second": 4.411,
      "step": 5733
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 0.6932,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930968761444092,
      "eval_runtime": 9.8362,
      "eval_samples_per_second": 555.395,
      "eval_steps_per_second": 4.372,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "step": 6552,
      "total_flos": 2.159818579325747e+16,
      "train_loss": 0.6932040762988639,
      "train_runtime": 4510.6708,
      "train_samples_per_second": 1161.058,
      "train_steps_per_second": 9.078
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 2.159818579325747e+16,
  "trial_name": null,
  "trial_params": null
}