{
  "best_metric": 0.541095890410959,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-96",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 480,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4191925525665283,
      "learning_rate": 0.00022738674133918383,
      "loss": 0.098,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.541095890410959,
      "eval_f1": 0.20642978003384094,
      "eval_loss": 0.08223184198141098,
      "eval_precision": 0.7625,
      "eval_recall": 0.11937377690802348,
      "eval_runtime": 28.0826,
      "eval_samples_per_second": 36.393,
      "eval_steps_per_second": 1.139,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.35244929790496826,
      "learning_rate": 0.00017054005600438787,
      "loss": 0.0833,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5371819960861057,
      "eval_f1": 0.19145299145299144,
      "eval_loss": 0.07941487431526184,
      "eval_precision": 0.7567567567567568,
      "eval_recall": 0.1095890410958904,
      "eval_runtime": 28.0455,
      "eval_samples_per_second": 36.441,
      "eval_steps_per_second": 1.141,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.2497691810131073,
      "learning_rate": 0.00011369337066959191,
      "loss": 0.0813,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5342465753424658,
      "eval_f1": 0.18213058419243985,
      "eval_loss": 0.07839526236057281,
      "eval_precision": 0.7464788732394366,
      "eval_recall": 0.10371819960861056,
      "eval_runtime": 28.0524,
      "eval_samples_per_second": 36.432,
      "eval_steps_per_second": 1.141,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.2048983871936798,
      "learning_rate": 5.684668533479596e-05,
      "loss": 0.0805,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5332681017612525,
      "eval_f1": 0.17331022530329288,
      "eval_loss": 0.07847526669502258,
      "eval_precision": 0.7575757575757576,
      "eval_recall": 0.09784735812133072,
      "eval_runtime": 28.003,
      "eval_samples_per_second": 36.496,
      "eval_steps_per_second": 1.143,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3579064607620239,
      "learning_rate": 0.0,
      "loss": 0.0801,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.538160469667319,
      "eval_f1": 0.1945392491467577,
      "eval_loss": 0.07800330221652985,
      "eval_precision": 0.76,
      "eval_recall": 0.11154598825831702,
      "eval_runtime": 27.9144,
      "eval_samples_per_second": 36.612,
      "eval_steps_per_second": 1.146,
      "step": 480
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1178475986400.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.09292461434001364,
    "learning_rate": 0.0002842334266739798,
    "num_train_epochs": 5,
    "temperature": 2
  }
}