trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.041666666666666664,
      "grad_norm": 73.5,
      "learning_rate": 0.0,
      "loss": 3.9287,
      "step": 1
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 80.5,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.994,
      "step": 5
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 63.75,
      "learning_rate": 9e-06,
      "loss": 3.6819,
      "step": 10
    },
    {
      "epoch": 0.625,
      "grad_norm": 49.25,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.5702,
      "step": 15
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 48.75,
      "learning_rate": 1.9e-05,
      "loss": 1.4286,
      "step": 20
    }
  ],
  "logging_steps": 5,
  "max_steps": 240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 63596453167104.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}