run-0/checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.1446540880503147,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6122580645161291,
      "eval_loss": 0.20235490798950195,
      "eval_runtime": 5.9992,
      "eval_samples_per_second": 516.736,
      "eval_steps_per_second": 10.835,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.550763701707098e-05,
      "loss": 0.3237,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8187096774193549,
      "eval_loss": 0.10080192238092422,
      "eval_runtime": 5.8671,
      "eval_samples_per_second": 528.366,
      "eval_steps_per_second": 11.079,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8764516129032258,
      "eval_loss": 0.06733943521976471,
      "eval_runtime": 5.745,
      "eval_samples_per_second": 539.6,
      "eval_steps_per_second": 11.314,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.101527403414196e-05,
      "loss": 0.1156,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2226,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 259991364709020.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7143131281575181,
    "num_train_epochs": 7,
    "temperature": 11
  }
}