trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.997289972899729,
  "eval_steps": 500,
  "global_step": 115,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17344173441734417,
      "grad_norm": 1.584091067314148,
      "learning_rate": 9.851888288072053e-05,
      "loss": 10.7111,
      "step": 20
    },
    {
      "epoch": 0.34688346883468835,
      "grad_norm": 0.5476922392845154,
      "learning_rate": 8.284768815411692e-05,
      "loss": 8.498,
      "step": 40
    },
    {
      "epoch": 0.5203252032520326,
      "grad_norm": 0.15487881004810333,
      "learning_rate": 5.532752499699381e-05,
      "loss": 7.8436,
      "step": 60
    },
    {
      "epoch": 0.6937669376693767,
      "grad_norm": 0.13645663857460022,
      "learning_rate": 2.588559263632719e-05,
      "loss": 7.7542,
      "step": 80
    },
    {
      "epoch": 0.8672086720867209,
      "grad_norm": 0.14501729607582092,
      "learning_rate": 5.1423196830513e-06,
      "loss": 7.6913,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 115,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.863515707905147e+17,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": null
}