{
  "best_metric": 0.7697715289982425,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-24/checkpoint-1284",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1284,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.816347122192383,
      "learning_rate": 0.0002514027907655722,
      "loss": 0.603,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.705916813122437,
      "eval_f1": 0.4670912951167728,
      "eval_loss": 0.5544006824493408,
      "eval_mcc": 0.2876964932013613,
      "eval_precision": 0.5898123324396782,
      "eval_recall": 0.3866432337434095,
      "eval_runtime": 3.1616,
      "eval_samples_per_second": 539.92,
      "eval_steps_per_second": 17.08,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.7939634323120117,
      "learning_rate": 0.00021997744191987566,
      "loss": 0.531,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7352079671939075,
      "eval_f1": 0.5368852459016393,
      "eval_loss": 0.5183561444282532,
      "eval_mcc": 0.3684302929196794,
      "eval_precision": 0.6437346437346437,
      "eval_recall": 0.46045694200351495,
      "eval_runtime": 3.2232,
      "eval_samples_per_second": 529.598,
      "eval_steps_per_second": 16.754,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.888092279434204,
      "learning_rate": 0.00018855209307417915,
      "loss": 0.502,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7451669595782073,
      "eval_f1": 0.48520710059171596,
      "eval_loss": 0.5155817270278931,
      "eval_mcc": 0.3814260429877007,
      "eval_precision": 0.7427536231884058,
      "eval_recall": 0.3602811950790861,
      "eval_runtime": 3.1891,
      "eval_samples_per_second": 535.256,
      "eval_steps_per_second": 16.933,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.3760223388671875,
      "learning_rate": 0.00015712674422848263,
      "loss": 0.4897,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7422378441710603,
      "eval_f1": 0.45812807881773393,
      "eval_loss": 0.5075758695602417,
      "eval_mcc": 0.37344089980427997,
      "eval_precision": 0.7654320987654321,
      "eval_recall": 0.3268892794376098,
      "eval_runtime": 3.1359,
      "eval_samples_per_second": 544.344,
      "eval_steps_per_second": 17.22,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.075637817382812,
      "learning_rate": 0.0001257013953827861,
      "loss": 0.4761,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7568834212067955,
      "eval_f1": 0.5870646766169154,
      "eval_loss": 0.49240508675575256,
      "eval_mcc": 0.42649635439740835,
      "eval_precision": 0.676605504587156,
      "eval_recall": 0.5184534270650264,
      "eval_runtime": 3.1156,
      "eval_samples_per_second": 547.891,
      "eval_steps_per_second": 17.332,
      "step": 1070
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.270783424377441,
      "learning_rate": 9.427604653708957e-05,
      "loss": 0.4671,
      "step": 1284
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7697715289982425,
      "eval_f1": 0.6288951841359773,
      "eval_loss": 0.48595207929611206,
      "eval_mcc": 0.4660789140327505,
      "eval_precision": 0.6795918367346939,
      "eval_recall": 0.5852372583479789,
      "eval_runtime": 3.1299,
      "eval_samples_per_second": 545.38,
      "eval_steps_per_second": 17.253,
      "step": 1284
    }
  ],
  "logging_steps": 500,
  "max_steps": 1926,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 3148653986640.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9932022812512611,
    "learning_rate": 0.0002828281396112687,
    "num_train_epochs": 9,
    "temperature": 23
  }
}