run-14/checkpoint-1054/trainer_state.json

{
  "best_metric": 0.9071100917431193,
  "best_model_checkpoint": "electra-distilled-sst\\run-14\\checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 17.319332122802734,
      "learning_rate": 4.8198886816999536e-05,
      "loss": 0.722,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9071100917431193,
      "eval_loss": 0.4791809618473053,
      "eval_runtime": 1.4471,
      "eval_samples_per_second": 602.582,
      "eval_steps_per_second": 4.837,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.41178894042969,
      "learning_rate": 4.131519756156356e-05,
      "loss": 0.3851,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9036697247706422,
      "eval_loss": 0.5533952116966248,
      "eval_runtime": 1.4377,
      "eval_samples_per_second": 606.531,
      "eval_steps_per_second": 4.869,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 4216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 373776249469380.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8898953307578259,
    "learning_rate": 5.5069514043487813e-05,
    "num_train_epochs": 8,
    "temperature": 23
  }
}