{
  "best_metric": 0.8258317025440313,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-26/checkpoint-260",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 260,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.020422458648682,
      "learning_rate": 0.00036726045612177425,
      "loss": 0.5512,
      "step": 52
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7397260273972602,
      "eval_f1": 0.7823240589198035,
      "eval_loss": 0.4979219436645508,
      "eval_precision": 0.6722925457102672,
      "eval_recall": 0.9354207436399217,
      "eval_runtime": 28.449,
      "eval_samples_per_second": 35.924,
      "eval_steps_per_second": 1.125,
      "step": 52
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.4581830501556396,
      "learning_rate": 0.00029380836489741936,
      "loss": 0.4888,
      "step": 104
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7788649706457925,
      "eval_f1": 0.8061749571183533,
      "eval_loss": 0.45355936884880066,
      "eval_precision": 0.7175572519083969,
      "eval_recall": 0.9197651663405088,
      "eval_runtime": 28.5368,
      "eval_samples_per_second": 35.813,
      "eval_steps_per_second": 1.121,
      "step": 104
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.563227415084839,
      "learning_rate": 0.00022035627367306455,
      "loss": 0.4608,
      "step": 156
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7759295499021527,
      "eval_f1": 0.8109000825763832,
      "eval_loss": 0.45456382632255554,
      "eval_precision": 0.7014285714285714,
      "eval_recall": 0.9608610567514677,
      "eval_runtime": 27.7846,
      "eval_samples_per_second": 36.783,
      "eval_steps_per_second": 1.152,
      "step": 156
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.4997444152832031,
      "learning_rate": 0.00014690418244870968,
      "loss": 0.4436,
      "step": 208
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8062622309197651,
      "eval_f1": 0.8176795580110496,
      "eval_loss": 0.4241831302642822,
      "eval_precision": 0.7721739130434783,
      "eval_recall": 0.8688845401174168,
      "eval_runtime": 28.0956,
      "eval_samples_per_second": 36.376,
      "eval_steps_per_second": 1.139,
      "step": 208
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.6734051704406738,
      "learning_rate": 7.345209122435484e-05,
      "loss": 0.4276,
      "step": 260
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8258317025440313,
      "eval_f1": 0.8405017921146953,
      "eval_loss": 0.41184401512145996,
      "eval_precision": 0.775206611570248,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 28.1973,
      "eval_samples_per_second": 36.245,
      "eval_steps_per_second": 1.135,
      "step": 260
    }
  ],
  "logging_steps": 500,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1178475986400.0,
  "train_batch_size": 59,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.834805680054491,
    "learning_rate": 0.0004407125473461291,
    "num_train_epochs": 6,
    "per_device_train_batch_size": 59,
    "temperature": 21
  }
}