run-2/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_loss": 0.2003006935119629,
      "eval_runtime": 3.3177,
      "eval_samples_per_second": 934.389,
      "eval_steps_per_second": 19.592,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.5493956804275513,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.3146,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.10496184974908829,
      "eval_runtime": 3.3448,
      "eval_samples_per_second": 926.816,
      "eval_steps_per_second": 19.433,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.07582292705774307,
      "eval_runtime": 3.4125,
      "eval_samples_per_second": 908.413,
      "eval_steps_per_second": 19.047,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5040295124053955,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.1211,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 1590,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 260941334653608.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.80288090122901,
    "num_train_epochs": 7,
    "temperature": 17
  }
}