{
  "best_metric": 0.8268101761252447,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-1728",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 1728,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8947777152061462,
      "learning_rate": 0.0005181908296523575,
      "loss": 0.4457,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7925636007827789,
      "eval_f1": 0.8090090090090091,
      "eval_loss": 0.41997030377388,
      "eval_mcc": 0.59400157079309,
      "eval_precision": 0.7495826377295493,
      "eval_recall": 0.8786692759295499,
      "eval_runtime": 66.1643,
      "eval_samples_per_second": 15.446,
      "eval_steps_per_second": 0.484,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.971238613128662,
      "learning_rate": 0.0004606140708020956,
      "loss": 0.4163,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7915851272015656,
      "eval_f1": 0.7957813998082454,
      "eval_loss": 0.39958736300468445,
      "eval_mcc": 0.5836633290639656,
      "eval_precision": 0.7800751879699248,
      "eval_recall": 0.812133072407045,
      "eval_runtime": 66.5614,
      "eval_samples_per_second": 15.354,
      "eval_steps_per_second": 0.481,
      "step": 384
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.333700656890869,
      "learning_rate": 0.00040303731195183364,
      "loss": 0.4018,
      "step": 576
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7915851272015656,
      "eval_f1": 0.7909715407262021,
      "eval_loss": 0.40145382285118103,
      "eval_mcc": 0.5831803046632951,
      "eval_precision": 0.7933070866141733,
      "eval_recall": 0.7886497064579256,
      "eval_runtime": 66.8237,
      "eval_samples_per_second": 15.294,
      "eval_steps_per_second": 0.479,
      "step": 576
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9171998500823975,
      "learning_rate": 0.00034546055310157167,
      "loss": 0.3976,
      "step": 768
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.786692759295499,
      "eval_f1": 0.783300198807157,
      "eval_loss": 0.40033411979675293,
      "eval_mcc": 0.5736667956847438,
      "eval_precision": 0.795959595959596,
      "eval_recall": 0.7710371819960861,
      "eval_runtime": 66.6803,
      "eval_samples_per_second": 15.327,
      "eval_steps_per_second": 0.48,
      "step": 768
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.3183633089065552,
      "learning_rate": 0.00028788379425130975,
      "loss": 0.3869,
      "step": 960
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8013698630136986,
      "eval_f1": 0.8031037827352084,
      "eval_loss": 0.39175575971603394,
      "eval_mcc": 0.60283323302071,
      "eval_precision": 0.7961538461538461,
      "eval_recall": 0.8101761252446184,
      "eval_runtime": 66.5306,
      "eval_samples_per_second": 15.361,
      "eval_steps_per_second": 0.481,
      "step": 960
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.770856499671936,
      "learning_rate": 0.0002303070354010478,
      "loss": 0.3843,
      "step": 1152
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8228962818003914,
      "eval_f1": 0.8350045578851412,
      "eval_loss": 0.38628602027893066,
      "eval_mcc": 0.6528627457969611,
      "eval_precision": 0.7815699658703071,
      "eval_recall": 0.8962818003913894,
      "eval_runtime": 67.0562,
      "eval_samples_per_second": 15.241,
      "eval_steps_per_second": 0.477,
      "step": 1152
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.1659635305404663,
      "learning_rate": 0.00017273027655078583,
      "loss": 0.3793,
      "step": 1344
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8111545988258317,
      "eval_f1": 0.8269058295964126,
      "eval_loss": 0.39157727360725403,
      "eval_mcc": 0.6328787442530197,
      "eval_precision": 0.7632450331125827,
      "eval_recall": 0.9021526418786693,
      "eval_runtime": 66.5334,
      "eval_samples_per_second": 15.361,
      "eval_steps_per_second": 0.481,
      "step": 1344
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.5721819400787354,
      "learning_rate": 0.0001151535177005239,
      "loss": 0.3727,
      "step": 1536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.821917808219178,
      "eval_f1": 0.8311688311688312,
      "eval_loss": 0.3822523355484009,
      "eval_mcc": 0.6477369472991997,
      "eval_precision": 0.7901234567901234,
      "eval_recall": 0.8767123287671232,
      "eval_runtime": 67.4301,
      "eval_samples_per_second": 15.156,
      "eval_steps_per_second": 0.475,
      "step": 1536
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.5437999963760376,
      "learning_rate": 5.757675885026195e-05,
      "loss": 0.3692,
      "step": 1728
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8268101761252447,
      "eval_f1": 0.8334901222953904,
      "eval_loss": 0.37901678681373596,
      "eval_mcc": 0.655734447578894,
      "eval_precision": 0.802536231884058,
      "eval_recall": 0.8669275929549902,
      "eval_runtime": 67.6441,
      "eval_samples_per_second": 15.108,
      "eval_steps_per_second": 0.473,
      "step": 1728
    }
  ],
  "logging_steps": 500,
  "max_steps": 1920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2121256775520.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6664545915670365,
    "learning_rate": 0.0005757675885026195,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 16,
    "temperature": 37
  }
}