checkpoint-39/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 12.16603946685791,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 3.235,
      "step": 10
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 2.4280266761779785,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 1.1738,
      "step": 20
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.5295767784118652,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.735,
      "step": 30
    }
  ],
  "logging_steps": 10,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1523313436262400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}