{
  "best_metric": 0.8055555555555556,
  "best_model_checkpoint": "test/checkpoint-6",
  "epoch": 19.727272727272727,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.73,
      "eval_accuracy": 0.6388888888888888,
      "eval_f1": 0.6477582846003898,
      "eval_loss": 0.679732084274292,
      "eval_precision": 0.6572420634920635,
      "eval_recall": 0.6388888888888888,
      "eval_runtime": 5.9427,
      "eval_samples_per_second": 6.058,
      "eval_steps_per_second": 0.168,
      "step": 2
    },
    {
      "epoch": 1.73,
      "eval_accuracy": 0.75,
      "eval_f1": 0.6904761904761904,
      "eval_loss": 0.6750033497810364,
      "eval_precision": 0.6397058823529411,
      "eval_recall": 0.75,
      "eval_runtime": 4.9005,
      "eval_samples_per_second": 7.346,
      "eval_steps_per_second": 0.204,
      "step": 4
    },
    {
      "epoch": 2.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.6533546447753906,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.9351,
      "eval_samples_per_second": 7.295,
      "eval_steps_per_second": 0.203,
      "step": 6
    },
    {
      "epoch": 3.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.6294739842414856,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.7154,
      "eval_samples_per_second": 7.635,
      "eval_steps_per_second": 0.212,
      "step": 8
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.0001,
      "loss": 0.8452,
      "step": 10
    },
    {
      "epoch": 4.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.6180709600448608,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.3103,
      "eval_samples_per_second": 6.779,
      "eval_steps_per_second": 0.188,
      "step": 10
    },
    {
      "epoch": 5.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.6019383668899536,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.7577,
      "eval_samples_per_second": 7.567,
      "eval_steps_per_second": 0.21,
      "step": 12
    },
    {
      "epoch": 6.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.6006457805633545,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.7756,
      "eval_samples_per_second": 7.538,
      "eval_steps_per_second": 0.209,
      "step": 14
    },
    {
      "epoch": 7.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.584420919418335,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.7897,
      "eval_samples_per_second": 6.218,
      "eval_steps_per_second": 0.173,
      "step": 16
    },
    {
      "epoch": 8.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5822530388832092,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.7388,
      "eval_samples_per_second": 7.597,
      "eval_steps_per_second": 0.211,
      "step": 18
    },
    {
      "epoch": 9.73,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.7824,
      "step": 20
    },
    {
      "epoch": 9.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5739169716835022,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.8104,
      "eval_samples_per_second": 7.484,
      "eval_steps_per_second": 0.208,
      "step": 20
    },
    {
      "epoch": 10.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5697076916694641,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.6974,
      "eval_samples_per_second": 7.664,
      "eval_steps_per_second": 0.213,
      "step": 22
    },
    {
      "epoch": 11.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5695454478263855,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.9353,
      "eval_samples_per_second": 7.294,
      "eval_steps_per_second": 0.203,
      "step": 24
    },
    {
      "epoch": 12.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5626423954963684,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.8521,
      "eval_samples_per_second": 6.152,
      "eval_steps_per_second": 0.171,
      "step": 26
    },
    {
      "epoch": 13.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5666201710700989,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.1217,
      "eval_samples_per_second": 7.029,
      "eval_steps_per_second": 0.195,
      "step": 28
    },
    {
      "epoch": 14.73,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.7149,
      "step": 30
    },
    {
      "epoch": 14.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5559208989143372,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.1084,
      "eval_samples_per_second": 7.047,
      "eval_steps_per_second": 0.196,
      "step": 30
    },
    {
      "epoch": 15.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5556015372276306,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 6.0749,
      "eval_samples_per_second": 5.926,
      "eval_steps_per_second": 0.165,
      "step": 32
    },
    {
      "epoch": 16.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5465428233146667,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.8659,
      "eval_samples_per_second": 7.398,
      "eval_steps_per_second": 0.206,
      "step": 34
    },
    {
      "epoch": 17.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5558464527130127,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.8604,
      "eval_samples_per_second": 7.407,
      "eval_steps_per_second": 0.206,
      "step": 36
    },
    {
      "epoch": 18.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5543624758720398,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 4.7553,
      "eval_samples_per_second": 7.57,
      "eval_steps_per_second": 0.21,
      "step": 38
    },
    {
      "epoch": 19.73,
      "learning_rate": 0.0,
      "loss": 0.6869,
      "step": 40
    },
    {
      "epoch": 19.73,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 0.5504012107849121,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 5.3116,
      "eval_samples_per_second": 6.778,
      "eval_steps_per_second": 0.188,
      "step": 40
    },
    {
      "epoch": 19.73,
      "step": 40,
      "total_flos": 2.8684782950326272e+17,
      "train_loss": 0.7573485970497131,
      "train_runtime": 850.4827,
      "train_samples_per_second": 16.085,
      "train_steps_per_second": 0.047
    }
  ],
  "max_steps": 40,
  "num_train_epochs": 20,
  "total_flos": 2.8684782950326272e+17,
  "trial_name": null,
  "trial_params": null
}