trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 0,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 4.834155559539795,
      "learning_rate": 1e-05,
      "loss": 0.9092,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.137935638427734,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.9497,
      "step": 2
    },
    {
      "epoch": 0.6,
      "grad_norm": 4.987179279327393,
      "learning_rate": 9.567727288213005e-06,
      "loss": 0.7857,
      "step": 3
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.295909881591797,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.7714,
      "step": 4
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.681474208831787,
      "learning_rate": 8.345653031794292e-06,
      "loss": 0.6515,
      "step": 5
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.703080892562866,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.56,
      "step": 6
    },
    {
      "epoch": 1.4,
      "grad_norm": 2.7864272594451904,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.5379,
      "step": 7
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.8693835735321045,
      "learning_rate": 5.522642316338268e-06,
      "loss": 0.509,
      "step": 8
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.8035385608673096,
      "learning_rate": 4.477357683661734e-06,
      "loss": 0.5536,
      "step": 9
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4770323038101196,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.4764,
      "step": 10
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.537730097770691,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.5115,
      "step": 11
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.3944271802902222,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.4966,
      "step": 12
    },
    {
      "epoch": 2.6,
      "grad_norm": 1.2273967266082764,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.4532,
      "step": 13
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.2184977531433105,
      "learning_rate": 4.322727117869951e-07,
      "loss": 0.4331,
      "step": 14
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.390999436378479,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 0.4964,
      "step": 15
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1758622325407744.0,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}