checkpoint-628/trainer_state.json

{
  "best_metric": 1.8257514238357544,
  "best_model_checkpoint": "gen_test/checkpoint-628",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 628,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 1.8853687047958374,
      "eval_runtime": 56.3363,
      "eval_samples_per_second": 177.505,
      "eval_steps_per_second": 2.787,
      "step": 157
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.842081069946289,
      "eval_runtime": 56.186,
      "eval_samples_per_second": 177.98,
      "eval_steps_per_second": 2.794,
      "step": 314
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.8316441774368286,
      "eval_runtime": 56.2573,
      "eval_samples_per_second": 177.755,
      "eval_steps_per_second": 2.791,
      "step": 471
    },
    {
      "epoch": 3.18,
      "grad_norm": 270056.0,
      "learning_rate": 4.076433121019109e-06,
      "loss": 1.8572,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.8257514238357544,
      "eval_runtime": 56.1934,
      "eval_samples_per_second": 177.957,
      "eval_steps_per_second": 2.794,
      "step": 628
    }
  ],
  "logging_steps": 500,
  "max_steps": 628,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 2.090336256e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}