checkpoint-100/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0125,
      "grad_norm": 0.08136817067861557,
      "learning_rate": 0.0004,
      "loss": 1.158,
      "step": 25
    },
    {
      "epoch": 0.025,
      "grad_norm": 0.0653829574584961,
      "learning_rate": 0.0004998852503731983,
      "loss": 1.0957,
      "step": 50
    },
    {
      "epoch": 0.0375,
      "grad_norm": 0.11592712253332138,
      "learning_rate": 0.0004993848168027977,
      "loss": 0.9276,
      "step": 75
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.08926476538181305,
      "learning_rate": 0.0004984880506341147,
      "loss": 1.0337,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.9111642837524414,
      "eval_runtime": 843.8058,
      "eval_samples_per_second": 1.3,
      "eval_steps_per_second": 0.021,
      "step": 100
    }
  ],
  "logging_steps": 25,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "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": 1.0051773955833856e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}