trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9989786886362089,
  "eval_steps": 500,
  "global_step": 917,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5446993940219241,
      "grad_norm": 285544.53125,
      "learning_rate": 5e-05,
      "loss": 19.9795,
      "step": 500
    },
    {
      "epoch": 0.5446993940219241,
      "eval_loss": 0.9794354438781738,
      "eval_runtime": 2716.2201,
      "eval_samples_per_second": 169.083,
      "eval_steps_per_second": 0.661,
      "step": 500
    },
    {
      "epoch": 0.9989786886362089,
      "step": 917,
      "total_flos": 6.402304001609564e+17,
      "train_loss": 17.88196479689204,
      "train_runtime": 28909.2161,
      "train_samples_per_second": 65.026,
      "train_steps_per_second": 0.032
    }
  ],
  "logging_steps": 500,
  "max_steps": 917,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 6.402304001609564e+17,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}