last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.10416666666666667,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003472222222222222,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.003472222222222222,
      "eval_loss": NaN,
      "eval_runtime": 115.1877,
      "eval_samples_per_second": 4.211,
      "eval_steps_per_second": 2.11,
      "step": 1
    },
    {
      "epoch": 0.006944444444444444,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.010416666666666666,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.013888888888888888,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.017361111111111112,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.020833333333333332,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.024305555555555556,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.027777777777777776,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.027777777777777776,
      "eval_loss": NaN,
      "eval_runtime": 19.9147,
      "eval_samples_per_second": 24.354,
      "eval_steps_per_second": 12.202,
      "step": 8
    },
    {
      "epoch": 0.03125,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.034722222222222224,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.03819444444444445,
      "grad_norm": NaN,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.041666666666666664,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.04513888888888889,
      "grad_norm": NaN,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.04861111111111111,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.052083333333333336,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.05555555555555555,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.05555555555555555,
      "eval_loss": NaN,
      "eval_runtime": 19.9217,
      "eval_samples_per_second": 24.345,
      "eval_steps_per_second": 12.198,
      "step": 16
    },
    {
      "epoch": 0.059027777777777776,
      "grad_norm": NaN,
      "learning_rate": 7.269952498697734e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0625,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.06597222222222222,
      "grad_norm": NaN,
      "learning_rate": 5.782172325201155e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.06944444444444445,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.07291666666666667,
      "grad_norm": NaN,
      "learning_rate": 4.2178276747988446e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0763888888888889,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.0798611111111111,
      "grad_norm": NaN,
      "learning_rate": 2.7300475013022663e-05,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.08333333333333333,
      "grad_norm": NaN,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.08333333333333333,
      "eval_loss": NaN,
      "eval_runtime": 19.9272,
      "eval_samples_per_second": 24.339,
      "eval_steps_per_second": 12.194,
      "step": 24
    },
    {
      "epoch": 0.08680555555555555,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.09027777777777778,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 26
    },
    {
      "epoch": 0.09375,
      "grad_norm": NaN,
      "learning_rate": 5.449673790581611e-06,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.09722222222222222,
      "grad_norm": NaN,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0,
      "step": 28
    },
    {
      "epoch": 0.10069444444444445,
      "grad_norm": NaN,
      "learning_rate": 6.15582970243117e-07,
      "loss": 0.0,
      "step": 29
    },
    {
      "epoch": 0.10416666666666667,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    }
  ],
  "logging_steps": 1,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7628045628211200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}