checkpoint-200/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.816901408450704,
  "eval_steps": 100,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.408450704225352,
      "grad_norm": 0.1609533429145813,
      "learning_rate": 3.766452673911396e-05,
      "loss": 0.2132,
      "step": 100
    },
    {
      "epoch": 1.408450704225352,
      "eval_loss": 0.050813838839530945,
      "eval_runtime": 130.758,
      "eval_samples_per_second": 4.344,
      "eval_steps_per_second": 2.172,
      "step": 100
    },
    {
      "epoch": 2.816901408450704,
      "grad_norm": 0.1740226149559021,
      "learning_rate": 1.0654091222300564e-05,
      "loss": 0.0432,
      "step": 200
    },
    {
      "epoch": 2.816901408450704,
      "eval_loss": 0.043040867894887924,
      "eval_runtime": 130.7758,
      "eval_samples_per_second": 4.343,
      "eval_steps_per_second": 2.172,
      "step": 200
    }
  ],
  "logging_steps": 100,
  "max_steps": 284,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.4207760867436544e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}