{
  "best_metric": 0.16314448416233063,
  "best_model_checkpoint": "../results/distilbert-base-uncased-agnews/run-31/checkpoint-6750",
  "epoch": 2.0,
  "global_step": 6750,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "learning_rate": 1.5e-05,
      "loss": 0.6831,
      "step": 500
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.27,
      "step": 1000
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.7391304347826085e-05,
      "loss": 0.2275,
      "step": 1500
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.4782608695652175e-05,
      "loss": 0.2233,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.217391304347826e-05,
      "loss": 0.2019,
      "step": 2500
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.956521739130435e-05,
      "loss": 0.1916,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9411666666666667,
      "eval_loss": 0.17406368255615234,
      "eval_runtime": 17.0218,
      "eval_samples_per_second": 704.977,
      "eval_steps_per_second": 88.122,
      "step": 3375
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.6956521739130433e-05,
      "loss": 0.1747,
      "step": 3500
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.4347826086956522e-05,
      "loss": 0.1382,
      "step": 4000
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.173913043478261e-05,
      "loss": 0.1389,
      "step": 4500
    },
    {
      "epoch": 1.48,
      "learning_rate": 9.130434782608697e-06,
      "loss": 0.1269,
      "step": 5000
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.521739130434783e-06,
      "loss": 0.1299,
      "step": 5500
    },
    {
      "epoch": 1.78,
      "learning_rate": 3.913043478260869e-06,
      "loss": 0.1328,
      "step": 6000
    },
    {
      "epoch": 1.93,
      "learning_rate": 1.3043478260869566e-06,
      "loss": 0.123,
      "step": 6500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9483333333333334,
      "eval_loss": 0.16314448416233063,
      "eval_runtime": 16.9602,
      "eval_samples_per_second": 707.537,
      "eval_steps_per_second": 88.442,
      "step": 6750
    },
    {
      "epoch": 2.0,
      "step": 6750,
      "total_flos": 1.0849763309273088e+16,
      "train_loss": 0.20914978592484085,
      "train_runtime": 832.6813,
      "train_samples_per_second": 259.403,
      "train_steps_per_second": 8.106
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9483333333333334,
      "eval_loss": 0.16314448416233063,
      "eval_runtime": 17.4366,
      "eval_samples_per_second": 688.209,
      "eval_steps_per_second": 86.026,
      "step": 6750
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.16520710289478302,
      "eval_runtime": 10.786,
      "eval_samples_per_second": 704.618,
      "eval_steps_per_second": 88.077,
      "step": 6750
    }
  ],
  "max_steps": 6750,
  "num_train_epochs": 2,
  "total_flos": 1.0849763309273088e+16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 32,
    "warmup_steps": 1000
  }
}