{
  "best_metric": 0.7521968365553603,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-31/checkpoint-1070",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1070,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.190187454223633,
      "learning_rate": 0.00012111783530103911,
      "loss": 0.6036,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7106033977738723,
      "eval_f1": 0.46304347826086956,
      "eval_loss": 0.5476045608520508,
      "eval_mcc": 0.2951848655219638,
      "eval_precision": 0.6068376068376068,
      "eval_recall": 0.37434094903339193,
      "eval_runtime": 3.1382,
      "eval_samples_per_second": 543.94,
      "eval_steps_per_second": 17.207,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.7711777687072754,
      "learning_rate": 0.0001009315294175326,
      "loss": 0.5367,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7305213825424721,
      "eval_f1": 0.5168067226890756,
      "eval_loss": 0.5195680260658264,
      "eval_mcc": 0.35250881017307234,
      "eval_precision": 0.6422976501305483,
      "eval_recall": 0.43233743409490333,
      "eval_runtime": 3.1515,
      "eval_samples_per_second": 541.653,
      "eval_steps_per_second": 17.135,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.508857488632202,
      "learning_rate": 8.074522353402607e-05,
      "loss": 0.5095,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7428236672524897,
      "eval_f1": 0.5284640171858218,
      "eval_loss": 0.5116792321205139,
      "eval_mcc": 0.38102846368132237,
      "eval_precision": 0.6795580110497238,
      "eval_recall": 0.43233743409490333,
      "eval_runtime": 3.1425,
      "eval_samples_per_second": 543.207,
      "eval_steps_per_second": 17.184,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.153243064880371,
      "learning_rate": 6.0558917650519556e-05,
      "loss": 0.5003,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7469244288224957,
      "eval_f1": 0.5364806866952789,
      "eval_loss": 0.5013575553894043,
      "eval_mcc": 0.3917806841670766,
      "eval_precision": 0.6887052341597796,
      "eval_recall": 0.43936731107205623,
      "eval_runtime": 3.159,
      "eval_samples_per_second": 540.365,
      "eval_steps_per_second": 17.094,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.839134693145752,
      "learning_rate": 4.037261176701304e-05,
      "loss": 0.4897,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7521968365553603,
      "eval_f1": 0.5952153110047848,
      "eval_loss": 0.49949556589126587,
      "eval_mcc": 0.42215185251347925,
      "eval_precision": 0.6533613445378151,
      "eval_recall": 0.546572934973638,
      "eval_runtime": 3.1467,
      "eval_samples_per_second": 542.465,
      "eval_steps_per_second": 17.161,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 1498,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 2623878322200.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9889630226398882,
    "learning_rate": 0.00014130414118454564,
    "num_train_epochs": 7,
    "temperature": 15
  }
}