{
  "best_metric": 0.7568627450980392,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/checkpoint-800",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 960,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9096550941467285,
      "learning_rate": 0.00038222682011154285,
      "loss": 0.5297,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.503944993019104,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8592,
      "eval_samples_per_second": 685.796,
      "eval_steps_per_second": 21.515,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1312909126281738,
      "learning_rate": 0.0003185223500929524,
      "loss": 0.496,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7396078431372549,
      "eval_f1": 0.5388888888888889,
      "eval_loss": 0.4670221507549286,
      "eval_mcc": 0.3774355933383126,
      "eval_precision": 0.6576271186440678,
      "eval_recall": 0.45647058823529413,
      "eval_runtime": 2.191,
      "eval_samples_per_second": 581.93,
      "eval_steps_per_second": 18.257,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.6883387565612793,
      "learning_rate": 0.0002548178800743619,
      "loss": 0.4793,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7388235294117647,
      "eval_f1": 0.5138686131386861,
      "eval_loss": 0.4671221673488617,
      "eval_mcc": 0.36889273398279,
      "eval_precision": 0.676923076923077,
      "eval_recall": 0.41411764705882353,
      "eval_runtime": 1.8737,
      "eval_samples_per_second": 680.471,
      "eval_steps_per_second": 21.348,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.2410541772842407,
      "learning_rate": 0.00019111341005577143,
      "loss": 0.4728,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7270588235294118,
      "eval_f1": 0.42,
      "eval_loss": 0.4601285755634308,
      "eval_mcc": 0.3271641233943044,
      "eval_precision": 0.72,
      "eval_recall": 0.2964705882352941,
      "eval_runtime": 2.208,
      "eval_samples_per_second": 577.447,
      "eval_steps_per_second": 18.116,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.100618362426758,
      "learning_rate": 0.00012740894003718096,
      "loss": 0.4617,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7568627450980392,
      "eval_f1": 0.5730027548209367,
      "eval_loss": 0.45551422238349915,
      "eval_mcc": 0.4218176989662163,
      "eval_precision": 0.6910299003322259,
      "eval_recall": 0.4894117647058824,
      "eval_runtime": 1.8811,
      "eval_samples_per_second": 677.808,
      "eval_steps_per_second": 21.265,
      "step": 800
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.64212965965271,
      "learning_rate": 6.370447001859048e-05,
      "loss": 0.4528,
      "step": 960
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.756078431372549,
      "eval_f1": 0.5674547983310153,
      "eval_loss": 0.45275548100471497,
      "eval_mcc": 0.41870095673277924,
      "eval_precision": 0.6938775510204082,
      "eval_recall": 0.48,
      "eval_runtime": 1.9208,
      "eval_samples_per_second": 663.796,
      "eval_steps_per_second": 20.825,
      "step": 960
    }
  ],
  "logging_steps": 500,
  "max_steps": 1120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1750532627520.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8351933279450521,
    "learning_rate": 0.00044593129013013335,
    "num_train_epochs": 7,
    "temperature": 12
  }
}