{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.0,
  "global_step": 11231,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49,
      "learning_rate": 1.9347045380346067e-05,
      "loss": 0.1628,
      "step": 500
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.8694090760692135e-05,
      "loss": 0.1424,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9604628682136536,
      "eval_loss": 0.13337615132331848,
      "eval_runtime": 3.606,
      "eval_samples_per_second": 2588.168,
      "eval_steps_per_second": 80.976,
      "step": 1021
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.80411361410382e-05,
      "loss": 0.1054,
      "step": 1500
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.7388181521384265e-05,
      "loss": 0.1097,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9652844667434692,
      "eval_loss": 0.11623150110244751,
      "eval_runtime": 3.5866,
      "eval_samples_per_second": 2602.181,
      "eval_steps_per_second": 81.414,
      "step": 2042
    },
    {
      "epoch": 2.45,
      "learning_rate": 1.673522690173033e-05,
      "loss": 0.0804,
      "step": 2500
    },
    {
      "epoch": 2.94,
      "learning_rate": 1.6082272282076395e-05,
      "loss": 0.0755,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9673202633857727,
      "eval_loss": 0.1282455027103424,
      "eval_runtime": 3.6184,
      "eval_samples_per_second": 2579.339,
      "eval_steps_per_second": 80.699,
      "step": 3063
    },
    {
      "epoch": 3.43,
      "learning_rate": 1.5429317662422463e-05,
      "loss": 0.0591,
      "step": 3500
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.4776363042768528e-05,
      "loss": 0.0494,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9676417112350464,
      "eval_loss": 0.14640793204307556,
      "eval_runtime": 3.593,
      "eval_samples_per_second": 2597.544,
      "eval_steps_per_second": 81.269,
      "step": 4084
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.4123408423114595e-05,
      "loss": 0.0328,
      "step": 4500
    },
    {
      "epoch": 4.9,
      "learning_rate": 1.3470453803460662e-05,
      "loss": 0.0353,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9633558392524719,
      "eval_loss": 0.19047445058822632,
      "eval_runtime": 3.6111,
      "eval_samples_per_second": 2584.563,
      "eval_steps_per_second": 80.863,
      "step": 5105
    },
    {
      "epoch": 5.39,
      "learning_rate": 1.2817499183806728e-05,
      "loss": 0.0209,
      "step": 5500
    },
    {
      "epoch": 5.88,
      "learning_rate": 1.2164544564152792e-05,
      "loss": 0.0182,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.968820333480835,
      "eval_loss": 0.20608273148536682,
      "eval_runtime": 3.6122,
      "eval_samples_per_second": 2583.731,
      "eval_steps_per_second": 80.837,
      "step": 6126
    },
    {
      "epoch": 6.37,
      "learning_rate": 1.1511589944498858e-05,
      "loss": 0.0144,
      "step": 6500
    },
    {
      "epoch": 6.86,
      "learning_rate": 1.0858635324844923e-05,
      "loss": 0.0126,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9689274430274963,
      "eval_loss": 0.24051280319690704,
      "eval_runtime": 3.603,
      "eval_samples_per_second": 2590.356,
      "eval_steps_per_second": 81.044,
      "step": 7147
    },
    {
      "epoch": 7.35,
      "learning_rate": 1.020568070519099e-05,
      "loss": 0.0086,
      "step": 7500
    },
    {
      "epoch": 7.84,
      "learning_rate": 9.552726085537057e-06,
      "loss": 0.0087,
      "step": 8000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9695703387260437,
      "eval_loss": 0.259203165769577,
      "eval_runtime": 3.6023,
      "eval_samples_per_second": 2590.86,
      "eval_steps_per_second": 81.06,
      "step": 8168
    },
    {
      "epoch": 8.33,
      "learning_rate": 8.899771465883122e-06,
      "loss": 0.0056,
      "step": 8500
    },
    {
      "epoch": 8.81,
      "learning_rate": 8.246816846229188e-06,
      "loss": 0.0068,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9706417918205261,
      "eval_loss": 0.2531001567840576,
      "eval_runtime": 3.6041,
      "eval_samples_per_second": 2589.528,
      "eval_steps_per_second": 81.018,
      "step": 9189
    },
    {
      "epoch": 9.3,
      "learning_rate": 7.593862226575253e-06,
      "loss": 0.0041,
      "step": 9500
    },
    {
      "epoch": 9.79,
      "learning_rate": 6.94090760692132e-06,
      "loss": 0.0034,
      "step": 10000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9699989557266235,
      "eval_loss": 0.281872034072876,
      "eval_runtime": 3.597,
      "eval_samples_per_second": 2594.647,
      "eval_steps_per_second": 81.178,
      "step": 10210
    },
    {
      "epoch": 10.28,
      "learning_rate": 6.287952987267385e-06,
      "loss": 0.0034,
      "step": 10500
    },
    {
      "epoch": 10.77,
      "learning_rate": 5.634998367613451e-06,
      "loss": 0.0032,
      "step": 11000
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9728918671607971,
      "eval_loss": 0.27420875430107117,
      "eval_runtime": 3.6954,
      "eval_samples_per_second": 2525.56,
      "eval_steps_per_second": 79.017,
      "step": 11231
    }
  ],
  "max_steps": 15315,
  "num_train_epochs": 15,
  "total_flos": 2.363421365580595e+16,
  "trial_name": null,
  "trial_params": null
}