{
  "best_metric": 0.8307240704500979,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-7660",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 7660,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.275282621383667,
      "learning_rate": 0.00013397987335151483,
      "loss": 0.5034,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7886497064579256,
      "eval_f1": 0.8153846153846154,
      "eval_loss": 0.4397984445095062,
      "eval_precision": 0.723823975720789,
      "eval_recall": 0.9334637964774951,
      "eval_runtime": 133.4478,
      "eval_samples_per_second": 7.658,
      "eval_steps_per_second": 1.918,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.074667453765869,
      "learning_rate": 0.00011909322075690206,
      "loss": 0.442,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8101761252446184,
      "eval_f1": 0.831889081455806,
      "eval_loss": 0.43401414155960083,
      "eval_precision": 0.7465007776049767,
      "eval_recall": 0.9393346379647749,
      "eval_runtime": 133.9782,
      "eval_samples_per_second": 7.628,
      "eval_steps_per_second": 1.911,
      "step": 1532
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.7122271060943604,
      "learning_rate": 0.0001042065681622893,
      "loss": 0.4346,
      "step": 2298
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.8250460405156538,
      "eval_loss": 0.4545851945877075,
      "eval_precision": 0.7791304347826087,
      "eval_recall": 0.8767123287671232,
      "eval_runtime": 131.1842,
      "eval_samples_per_second": 7.791,
      "eval_steps_per_second": 1.951,
      "step": 2298
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.8108601570129395,
      "learning_rate": 8.931991556767654e-05,
      "loss": 0.4272,
      "step": 3064
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8258317025440313,
      "eval_f1": 0.8372943327239488,
      "eval_loss": 0.4149971902370453,
      "eval_precision": 0.7855917667238422,
      "eval_recall": 0.8962818003913894,
      "eval_runtime": 131.3897,
      "eval_samples_per_second": 7.778,
      "eval_steps_per_second": 1.948,
      "step": 3064
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.14157485961914,
      "learning_rate": 7.443326297306379e-05,
      "loss": 0.4139,
      "step": 3830
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8228962818003914,
      "eval_f1": 0.8353048225659692,
      "eval_loss": 0.43950268626213074,
      "eval_precision": 0.7806122448979592,
      "eval_recall": 0.898238747553816,
      "eval_runtime": 131.4228,
      "eval_samples_per_second": 7.776,
      "eval_steps_per_second": 1.948,
      "step": 3830
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.7704387903213501,
      "learning_rate": 5.954661037845103e-05,
      "loss": 0.4072,
      "step": 4596
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8180039138943248,
      "eval_f1": 0.8318264014466545,
      "eval_loss": 0.4495176076889038,
      "eval_precision": 0.773109243697479,
      "eval_recall": 0.9001956947162426,
      "eval_runtime": 132.0859,
      "eval_samples_per_second": 7.737,
      "eval_steps_per_second": 1.938,
      "step": 4596
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.204219818115234,
      "learning_rate": 4.465995778383827e-05,
      "loss": 0.3994,
      "step": 5362
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8238747553816047,
      "eval_f1": 0.8372513562386981,
      "eval_loss": 0.4403352737426758,
      "eval_precision": 0.7781512605042017,
      "eval_recall": 0.9060665362035225,
      "eval_runtime": 133.4911,
      "eval_samples_per_second": 7.656,
      "eval_steps_per_second": 1.918,
      "step": 5362
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.954471588134766,
      "learning_rate": 2.9773305189225516e-05,
      "loss": 0.392,
      "step": 6128
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8268101761252447,
      "eval_f1": 0.8395285584768812,
      "eval_loss": 0.43856292963027954,
      "eval_precision": 0.7820945945945946,
      "eval_recall": 0.9060665362035225,
      "eval_runtime": 131.9827,
      "eval_samples_per_second": 7.743,
      "eval_steps_per_second": 1.94,
      "step": 6128
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.39264997839927673,
      "learning_rate": 1.4886652594612758e-05,
      "loss": 0.3917,
      "step": 6894
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8287671232876712,
      "eval_f1": 0.8353715898400752,
      "eval_loss": 0.4313138425350189,
      "eval_precision": 0.8043478260869565,
      "eval_recall": 0.8688845401174168,
      "eval_runtime": 132.2212,
      "eval_samples_per_second": 7.729,
      "eval_steps_per_second": 1.936,
      "step": 6894
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.9275560975074768,
      "learning_rate": 0.0,
      "loss": 0.3787,
      "step": 7660
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8307240704500979,
      "eval_f1": 0.840258541089566,
      "eval_loss": 0.4347919523715973,
      "eval_precision": 0.7954545454545454,
      "eval_recall": 0.8904109589041096,
      "eval_runtime": 133.1343,
      "eval_samples_per_second": 7.676,
      "eval_steps_per_second": 1.923,
      "step": 7660
    }
  ],
  "logging_steps": 500,
  "max_steps": 7660,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2356951972800.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8157328229875264,
    "learning_rate": 0.00014886652594612757,
    "num_train_epochs": 10,
    "temperature": 20
  }
}