{
  "best_metric": 0.8284272507637513,
  "best_model_checkpoint": "test-klue/ynat/run-1/checkpoint-500",
  "epoch": 3.4965034965034967,
  "eval_steps": 50,
  "global_step": 500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35,
      "learning_rate": 5.859444109925402e-05,
      "loss": 1.497,
      "step": 50
    },
    {
      "epoch": 0.35,
      "eval_f1": 0.6734208042191661,
      "eval_loss": 1.089546799659729,
      "eval_runtime": 13.0037,
      "eval_samples_per_second": 700.339,
      "eval_steps_per_second": 1.384,
      "step": 50
    },
    {
      "epoch": 0.7,
      "learning_rate": 5.418884402412214e-05,
      "loss": 0.5665,
      "step": 100
    },
    {
      "epoch": 0.7,
      "eval_f1": 0.8012954608718272,
      "eval_loss": 0.6758251190185547,
      "eval_runtime": 13.0152,
      "eval_samples_per_second": 699.72,
      "eval_steps_per_second": 1.383,
      "step": 100
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.9783246948990256e-05,
      "loss": 0.5083,
      "step": 150
    },
    {
      "epoch": 1.05,
      "eval_f1": 0.7895493614371508,
      "eval_loss": 0.6960522532463074,
      "eval_runtime": 12.9373,
      "eval_samples_per_second": 703.936,
      "eval_steps_per_second": 1.391,
      "step": 150
    },
    {
      "epoch": 1.4,
      "learning_rate": 4.537764987385838e-05,
      "loss": 0.3865,
      "step": 200
    },
    {
      "epoch": 1.4,
      "eval_f1": 0.8201909522421974,
      "eval_loss": 0.6086650490760803,
      "eval_runtime": 12.7238,
      "eval_samples_per_second": 715.746,
      "eval_steps_per_second": 1.415,
      "step": 200
    },
    {
      "epoch": 1.75,
      "learning_rate": 4.09720527987265e-05,
      "loss": 0.385,
      "step": 250
    },
    {
      "epoch": 1.75,
      "eval_f1": 0.8359538639496097,
      "eval_loss": 0.50257807970047,
      "eval_runtime": 12.8233,
      "eval_samples_per_second": 710.19,
      "eval_steps_per_second": 1.404,
      "step": 250
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.656645572359462e-05,
      "loss": 0.357,
      "step": 300
    },
    {
      "epoch": 2.1,
      "eval_f1": 0.8294020659039051,
      "eval_loss": 0.5500179529190063,
      "eval_runtime": 12.8836,
      "eval_samples_per_second": 706.867,
      "eval_steps_per_second": 1.397,
      "step": 300
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.216085864846273e-05,
      "loss": 0.2337,
      "step": 350
    },
    {
      "epoch": 2.45,
      "eval_f1": 0.8324393645838855,
      "eval_loss": 0.5851877927780151,
      "eval_runtime": 12.9246,
      "eval_samples_per_second": 704.625,
      "eval_steps_per_second": 1.393,
      "step": 350
    },
    {
      "epoch": 2.8,
      "learning_rate": 2.775526157333085e-05,
      "loss": 0.2136,
      "step": 400
    },
    {
      "epoch": 2.8,
      "eval_f1": 0.8449098012505971,
      "eval_loss": 0.5285059809684753,
      "eval_runtime": 12.7922,
      "eval_samples_per_second": 711.919,
      "eval_steps_per_second": 1.407,
      "step": 400
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.3349664498198968e-05,
      "loss": 0.1926,
      "step": 450
    },
    {
      "epoch": 3.15,
      "eval_f1": 0.8313597336118761,
      "eval_loss": 0.6143582463264465,
      "eval_runtime": 12.8169,
      "eval_samples_per_second": 710.548,
      "eval_steps_per_second": 1.404,
      "step": 450
    },
    {
      "epoch": 3.5,
      "learning_rate": 1.894406742306709e-05,
      "loss": 0.1359,
      "step": 500
    },
    {
      "epoch": 3.5,
      "eval_f1": 0.8284272507637513,
      "eval_loss": 0.7059531211853027,
      "eval_runtime": 12.7876,
      "eval_samples_per_second": 712.173,
      "eval_steps_per_second": 1.408,
      "step": 500
    }
  ],
  "logging_steps": 50,
  "max_steps": 715,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 178689205892400.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.859444109925402e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 32,
    "seed": 11
  }
}