trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 2460,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012195121951219513,
      "grad_norm": 32.47914123535156,
      "learning_rate": 5e-05,
      "loss": 2.2038,
      "step": 1
    },
    {
      "epoch": 6.097560975609756,
      "grad_norm": 0.004238747991621494,
      "learning_rate": 3.9857723577235775e-05,
      "loss": 0.0401,
      "step": 500
    },
    {
      "epoch": 12.195121951219512,
      "grad_norm": 0.0014152198564261198,
      "learning_rate": 2.969512195121951e-05,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 18.29268292682927,
      "grad_norm": 0.0011360333301126957,
      "learning_rate": 1.953252032520325e-05,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 24.390243902439025,
      "grad_norm": 0.000390420900657773,
      "learning_rate": 9.369918699186993e-06,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 30.0,
      "step": 2460,
      "total_flos": 7.832575533868646e+17,
      "train_loss": 0.009042149140476817,
      "train_runtime": 1040.7829,
      "train_samples_per_second": 150.81,
      "train_steps_per_second": 2.364
    }
  ],
  "logging_steps": 500,
  "max_steps": 2460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.832575533868646e+17,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}