trainer_state.json

{
  "best_metric": 0.6899110078811646,
  "best_model_checkpoint": "add_BERT_48_qnli/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": 3.9200000000000004e-05,
      "loss": 0.6959,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.510342302764049,
      "eval_loss": 0.6932767629623413,
      "eval_runtime": 10.5297,
      "eval_samples_per_second": 518.819,
      "eval_steps_per_second": 4.084,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.6945,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6917039155960083,
      "eval_runtime": 10.5307,
      "eval_samples_per_second": 518.767,
      "eval_steps_per_second": 4.083,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6938,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5262676185246201,
      "eval_loss": 0.6899110078811646,
      "eval_runtime": 10.5065,
      "eval_samples_per_second": 519.965,
      "eval_steps_per_second": 4.093,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6925,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5317591067179206,
      "eval_loss": 0.690322995185852,
      "eval_runtime": 10.5223,
      "eval_samples_per_second": 519.183,
      "eval_steps_per_second": 4.087,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6961,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5222405271828665,
      "eval_loss": 0.6940132975578308,
      "eval_runtime": 10.5251,
      "eval_samples_per_second": 519.046,
      "eval_steps_per_second": 4.085,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.6929,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5251693208859601,
      "eval_loss": 0.6913561224937439,
      "eval_runtime": 10.4977,
      "eval_samples_per_second": 520.398,
      "eval_steps_per_second": 4.096,
      "step": 4914
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.6922,
      "step": 5733
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5043016657514187,
      "eval_loss": 0.6920790076255798,
      "eval_runtime": 10.5273,
      "eval_samples_per_second": 518.937,
      "eval_steps_per_second": 4.085,
      "step": 5733
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.692,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5302947098663738,
      "eval_loss": 0.6903797388076782,
      "eval_runtime": 10.5361,
      "eval_samples_per_second": 518.503,
      "eval_steps_per_second": 4.081,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "step": 6552,
      "total_flos": 1.1967849060892672e+17,
      "train_loss": 0.6937320322635264,
      "train_runtime": 5209.16,
      "train_samples_per_second": 1005.373,
      "train_steps_per_second": 7.861
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 1.1967849060892672e+17,
  "trial_name": null,
  "trial_params": null
}