{
  "best_metric": 0.8307240704500979,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-41/checkpoint-672",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 768,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4870496988296509,
      "learning_rate": 0.00011097537892965708,
      "loss": 0.5464,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7749510763209393,
      "eval_f1": 0.8017241379310346,
      "eval_loss": 0.4673503041267395,
      "eval_precision": 0.7164869029275809,
      "eval_recall": 0.9099804305283757,
      "eval_runtime": 28.2108,
      "eval_samples_per_second": 36.227,
      "eval_steps_per_second": 1.134,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9109100103378296,
      "learning_rate": 9.710345656344995e-05,
      "loss": 0.4617,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8082191780821918,
      "eval_f1": 0.8262411347517731,
      "eval_loss": 0.43051373958587646,
      "eval_precision": 0.7552674230145867,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 28.0456,
      "eval_samples_per_second": 36.441,
      "eval_steps_per_second": 1.141,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.547697067260742,
      "learning_rate": 8.32315341972428e-05,
      "loss": 0.4341,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8228962818003914,
      "eval_f1": 0.8337924701561066,
      "eval_loss": 0.41541436314582825,
      "eval_precision": 0.7854671280276817,
      "eval_recall": 0.8884540117416829,
      "eval_runtime": 28.3605,
      "eval_samples_per_second": 36.036,
      "eval_steps_per_second": 1.128,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.3498454093933105,
      "learning_rate": 6.935961183103568e-05,
      "loss": 0.4236,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8170254403131115,
      "eval_f1": 0.8203650336215178,
      "eval_loss": 0.4138866066932678,
      "eval_precision": 0.8056603773584906,
      "eval_recall": 0.8356164383561644,
      "eval_runtime": 28.1596,
      "eval_samples_per_second": 36.293,
      "eval_steps_per_second": 1.136,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.9074805974960327,
      "learning_rate": 5.548768946482854e-05,
      "loss": 0.4117,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8287671232876712,
      "eval_f1": 0.8427672955974842,
      "eval_loss": 0.40257665514945984,
      "eval_precision": 0.7790697674418605,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 28.5426,
      "eval_samples_per_second": 35.806,
      "eval_steps_per_second": 1.121,
      "step": 480
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.1893153190612793,
      "learning_rate": 4.16157670986214e-05,
      "loss": 0.4039,
      "step": 576
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.824853228962818,
      "eval_f1": 0.8385933273219117,
      "eval_loss": 0.4039454162120819,
      "eval_precision": 0.7775919732441472,
      "eval_recall": 0.9099804305283757,
      "eval_runtime": 28.1064,
      "eval_samples_per_second": 36.362,
      "eval_steps_per_second": 1.139,
      "step": 576
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.368473768234253,
      "learning_rate": 2.774384473241427e-05,
      "loss": 0.3968,
      "step": 672
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8307240704500979,
      "eval_f1": 0.8464951197870452,
      "eval_loss": 0.4061815142631531,
      "eval_precision": 0.7743506493506493,
      "eval_recall": 0.9334637964774951,
      "eval_runtime": 28.8284,
      "eval_samples_per_second": 35.451,
      "eval_steps_per_second": 1.11,
      "step": 672
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.229116916656494,
      "learning_rate": 1.3871922366207134e-05,
      "loss": 0.3973,
      "step": 768
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8307240704500979,
      "eval_f1": 0.8434389140271493,
      "eval_loss": 0.3992971181869507,
      "eval_precision": 0.7845117845117845,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 28.7392,
      "eval_samples_per_second": 35.561,
      "eval_steps_per_second": 1.113,
      "step": 768
    }
  ],
  "logging_steps": 500,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 1885561578240.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8078288065121237,
    "learning_rate": 0.0001248473012958642,
    "num_train_epochs": 9,
    "temperature": 4
  }
}