trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9999786329351937,
  "eval_steps": 500,
  "global_step": 11700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08546825922523023,
      "grad_norm": 3.453125,
      "learning_rate": 1.910714285714286e-05,
      "loss": 1.0595,
      "step": 1000
    },
    {
      "epoch": 0.17093651845046046,
      "grad_norm": 2.25,
      "learning_rate": 1.7321428571428572e-05,
      "loss": 0.2969,
      "step": 2000
    },
    {
      "epoch": 0.2564047776756907,
      "grad_norm": 2.0625,
      "learning_rate": 1.553571428571429e-05,
      "loss": 0.2876,
      "step": 3000
    },
    {
      "epoch": 0.3418730369009209,
      "grad_norm": 2.0625,
      "learning_rate": 1.375e-05,
      "loss": 0.2886,
      "step": 4000
    },
    {
      "epoch": 0.4273412961261511,
      "grad_norm": 1.7578125,
      "learning_rate": 1.1964285714285716e-05,
      "loss": 0.2888,
      "step": 5000
    },
    {
      "epoch": 0.5128095553513814,
      "grad_norm": 3.4375,
      "learning_rate": 1.0178571428571429e-05,
      "loss": 0.2836,
      "step": 6000
    },
    {
      "epoch": 0.5982778145766117,
      "grad_norm": 2.09375,
      "learning_rate": 8.392857142857144e-06,
      "loss": 0.2827,
      "step": 7000
    },
    {
      "epoch": 0.6837460738018418,
      "grad_norm": 2.90625,
      "learning_rate": 6.607142857142858e-06,
      "loss": 0.2865,
      "step": 8000
    },
    {
      "epoch": 0.7692143330270721,
      "grad_norm": 2.828125,
      "learning_rate": 4.821428571428572e-06,
      "loss": 0.2839,
      "step": 9000
    },
    {
      "epoch": 0.8546825922523023,
      "grad_norm": 2.640625,
      "learning_rate": 3.0357142857142856e-06,
      "loss": 0.2846,
      "step": 10000
    },
    {
      "epoch": 0.9401508514775325,
      "grad_norm": 2.3125,
      "learning_rate": 1.25e-06,
      "loss": 0.2852,
      "step": 11000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 11700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.0868950925312e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}