{
  "best_global_step": 626,
  "best_metric": 0.7155844155844155,
  "best_model_checkpoint": "models/saved_model\\checkpoint-626",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 626,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 3.755857229232788,
      "learning_rate": 1.894568690095847e-05,
      "loss": 4.254554443359375,
      "step": 100
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 5.2450270652771,
      "learning_rate": 1.788072417465389e-05,
      "loss": 3.837513427734375,
      "step": 200
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 5.9568634033203125,
      "learning_rate": 1.6815761448349307e-05,
      "loss": 3.35986328125,
      "step": 300
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 5.645184516906738,
      "learning_rate": 1.5750798722044728e-05,
      "loss": 2.946711120605469,
      "step": 400
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 6.78502082824707,
      "learning_rate": 1.468583599574015e-05,
      "loss": 2.553294830322266,
      "step": 500
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 6.254422187805176,
      "learning_rate": 1.3620873269435571e-05,
      "loss": 2.2656430053710936,
      "step": 600
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7155844155844155,
      "eval_loss": 1.9922140836715698,
      "eval_runtime": 358.6864,
      "eval_samples_per_second": 8.587,
      "eval_steps_per_second": 0.538,
      "step": 626
    }
  ],
  "logging_steps": 100,
  "max_steps": 1878,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 165855462036096.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}