{
  "best_metric": 0.6536203522504892,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-864",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 864,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8228768110275269,
      "learning_rate": 1.922542063523116e-05,
      "loss": 0.7149,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.007766990291262136,
      "eval_loss": 0.7131175398826599,
      "eval_precision": 0.5,
      "eval_recall": 0.003913894324853229,
      "eval_runtime": 26.5316,
      "eval_samples_per_second": 38.52,
      "eval_steps_per_second": 1.206,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1306586265563965,
      "learning_rate": 1.7089262786872146e-05,
      "loss": 0.7062,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5136986301369864,
      "eval_f1": 0.08133086876155268,
      "eval_loss": 0.7069234848022461,
      "eval_precision": 0.7333333333333333,
      "eval_recall": 0.043052837573385516,
      "eval_runtime": 26.8847,
      "eval_samples_per_second": 38.014,
      "eval_steps_per_second": 1.19,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5551536083221436,
      "learning_rate": 1.4953104938513126e-05,
      "loss": 0.6972,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.512720156555773,
      "eval_f1": 0.08791208791208792,
      "eval_loss": 0.6998902559280396,
      "eval_precision": 0.6857142857142857,
      "eval_recall": 0.046966731898238745,
      "eval_runtime": 26.6683,
      "eval_samples_per_second": 38.323,
      "eval_steps_per_second": 1.2,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.080883741378784,
      "learning_rate": 1.2816947090154107e-05,
      "loss": 0.6854,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5606653620352251,
      "eval_f1": 0.3145038167938931,
      "eval_loss": 0.6807624697685242,
      "eval_precision": 0.7152777777777778,
      "eval_recall": 0.20156555772994128,
      "eval_runtime": 26.6572,
      "eval_samples_per_second": 38.339,
      "eval_steps_per_second": 1.2,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0045652389526367,
      "learning_rate": 1.068078924179509e-05,
      "loss": 0.6688,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6105675146771037,
      "eval_f1": 0.49234693877551017,
      "eval_loss": 0.6664517521858215,
      "eval_precision": 0.706959706959707,
      "eval_recall": 0.3776908023483366,
      "eval_runtime": 26.8808,
      "eval_samples_per_second": 38.02,
      "eval_steps_per_second": 1.19,
      "step": 480
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.2470343112945557,
      "learning_rate": 8.544631393436073e-06,
      "loss": 0.6549,
      "step": 576
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6418786692759295,
      "eval_f1": 0.5906040268456376,
      "eval_loss": 0.6583118438720703,
      "eval_precision": 0.6892950391644909,
      "eval_recall": 0.5166340508806262,
      "eval_runtime": 26.2817,
      "eval_samples_per_second": 38.886,
      "eval_steps_per_second": 1.218,
      "step": 576
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.420502185821533,
      "learning_rate": 6.4084735450770535e-06,
      "loss": 0.6502,
      "step": 672
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6360078277886497,
      "eval_f1": 0.5782312925170068,
      "eval_loss": 0.6568923592567444,
      "eval_precision": 0.6873315363881402,
      "eval_recall": 0.49902152641878667,
      "eval_runtime": 26.5443,
      "eval_samples_per_second": 38.502,
      "eval_steps_per_second": 1.206,
      "step": 672
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.7794277667999268,
      "learning_rate": 4.2723156967180365e-06,
      "loss": 0.6453,
      "step": 768
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.649706457925636,
      "eval_f1": 0.6108695652173913,
      "eval_loss": 0.6531394720077515,
      "eval_precision": 0.687041564792176,
      "eval_recall": 0.5499021526418787,
      "eval_runtime": 26.8208,
      "eval_samples_per_second": 38.105,
      "eval_steps_per_second": 1.193,
      "step": 768
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.565406560897827,
      "learning_rate": 2.1361578483590183e-06,
      "loss": 0.6432,
      "step": 864
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6536203522504892,
      "eval_f1": 0.616052060737527,
      "eval_loss": 0.6520226001739502,
      "eval_precision": 0.6909975669099757,
      "eval_recall": 0.5557729941291585,
      "eval_runtime": 26.3078,
      "eval_samples_per_second": 38.848,
      "eval_steps_per_second": 1.216,
      "step": 864
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1984401499680.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8789746577248793,
    "learning_rate": 2.136157848359018e-05,
    "num_train_epochs": 10,
    "temperature": 14
  }
}