{
  "best_metric": 0.7750439367311072,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-2140",
  "epoch": 11.0,
  "eval_steps": 500,
  "global_step": 2354,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.918547630310059,
      "learning_rate": 0.00018354791279082705,
      "loss": 0.5995,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7053309900410076,
      "eval_f1": 0.4944723618090453,
      "eval_loss": 0.5527746677398682,
      "eval_mcc": 0.29864842708933803,
      "eval_precision": 0.5774647887323944,
      "eval_recall": 0.43233743409490333,
      "eval_runtime": 3.1509,
      "eval_samples_per_second": 541.75,
      "eval_steps_per_second": 17.138,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.0788278579711914,
      "learning_rate": 0.00016519312151174438,
      "loss": 0.5413,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7258347978910369,
      "eval_f1": 0.4845814977973568,
      "eval_loss": 0.5296629071235657,
      "eval_mcc": 0.33330907379505725,
      "eval_precision": 0.6489675516224189,
      "eval_recall": 0.3866432337434095,
      "eval_runtime": 3.2724,
      "eval_samples_per_second": 521.631,
      "eval_steps_per_second": 16.502,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.197441339492798,
      "learning_rate": 0.00014683833023266166,
      "loss": 0.5154,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7258347978910369,
      "eval_f1": 0.38421052631578945,
      "eval_loss": 0.5396824479103088,
      "eval_mcc": 0.32457545244467173,
      "eval_precision": 0.7643979057591623,
      "eval_recall": 0.2565905096660808,
      "eval_runtime": 3.1857,
      "eval_samples_per_second": 535.83,
      "eval_steps_per_second": 16.951,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.9606778621673584,
      "learning_rate": 0.00012848353895357893,
      "loss": 0.504,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7264206209724663,
      "eval_f1": 0.3942931258106355,
      "eval_loss": 0.5155606269836426,
      "eval_mcc": 0.32574276921975914,
      "eval_precision": 0.7524752475247525,
      "eval_recall": 0.2671353251318102,
      "eval_runtime": 3.3181,
      "eval_samples_per_second": 514.447,
      "eval_steps_per_second": 16.274,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.5369954109191895,
      "learning_rate": 0.00011012874767449624,
      "loss": 0.4946,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7504393673110721,
      "eval_f1": 0.5590062111801243,
      "eval_loss": 0.5006516575813293,
      "eval_mcc": 0.4049525635550248,
      "eval_precision": 0.6801007556675063,
      "eval_recall": 0.47451669595782076,
      "eval_runtime": 3.155,
      "eval_samples_per_second": 541.038,
      "eval_steps_per_second": 17.115,
      "step": 1070
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.3806962966918945,
      "learning_rate": 9.177395639541353e-05,
      "loss": 0.4829,
      "step": 1284
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7586408904510837,
      "eval_f1": 0.603082851637765,
      "eval_loss": 0.486370325088501,
      "eval_mcc": 0.43615024061600965,
      "eval_precision": 0.6673773987206824,
      "eval_recall": 0.5500878734622144,
      "eval_runtime": 3.9493,
      "eval_samples_per_second": 432.226,
      "eval_steps_per_second": 13.673,
      "step": 1284
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.009945392608643,
      "learning_rate": 7.341916511633083e-05,
      "loss": 0.4704,
      "step": 1498
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7580550673696543,
      "eval_f1": 0.5100830367734281,
      "eval_loss": 0.4908476173877716,
      "eval_mcc": 0.41865909968733495,
      "eval_precision": 0.7846715328467153,
      "eval_recall": 0.37785588752196836,
      "eval_runtime": 3.1648,
      "eval_samples_per_second": 539.371,
      "eval_steps_per_second": 17.063,
      "step": 1498
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.42345666885376,
      "learning_rate": 5.506437383724812e-05,
      "loss": 0.4564,
      "step": 1712
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7662565905096661,
      "eval_f1": 0.5973763874873865,
      "eval_loss": 0.47338271141052246,
      "eval_mcc": 0.4474691814812865,
      "eval_precision": 0.7014218009478673,
      "eval_recall": 0.5202108963093146,
      "eval_runtime": 3.2515,
      "eval_samples_per_second": 524.99,
      "eval_steps_per_second": 16.608,
      "step": 1712
    },
    {
      "epoch": 9.0,
      "grad_norm": 7.744714260101318,
      "learning_rate": 3.6709582558165414e-05,
      "loss": 0.4528,
      "step": 1926
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7656707674282367,
      "eval_f1": 0.5975855130784709,
      "eval_loss": 0.4673592448234558,
      "eval_mcc": 0.4464084855010257,
      "eval_precision": 0.6988235294117647,
      "eval_recall": 0.5219683655536028,
      "eval_runtime": 3.1615,
      "eval_samples_per_second": 539.929,
      "eval_steps_per_second": 17.08,
      "step": 1926
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.320211887359619,
      "learning_rate": 1.8354791279082707e-05,
      "loss": 0.4449,
      "step": 2140
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7750439367311072,
      "eval_f1": 0.6335877862595419,
      "eval_loss": 0.4649231731891632,
      "eval_mcc": 0.47665988640122975,
      "eval_precision": 0.6931106471816284,
      "eval_recall": 0.5834797891036907,
      "eval_runtime": 3.2751,
      "eval_samples_per_second": 521.205,
      "eval_steps_per_second": 16.488,
      "step": 2140
    },
    {
      "epoch": 11.0,
      "grad_norm": 5.221016883850098,
      "learning_rate": 0.0,
      "loss": 0.4371,
      "step": 2354
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7738722905682484,
      "eval_f1": 0.6405959031657356,
      "eval_loss": 0.46391916275024414,
      "eval_mcc": 0.4782970414232871,
      "eval_precision": 0.6811881188118812,
      "eval_recall": 0.6045694200351494,
      "eval_runtime": 3.1542,
      "eval_samples_per_second": 541.178,
      "eval_steps_per_second": 17.12,
      "step": 2354
    }
  ],
  "logging_steps": 500,
  "max_steps": 2354,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 11,
  "save_steps": 500,
  "total_flos": 5772532308840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9896635045471636,
    "learning_rate": 0.00020190270406990977,
    "num_train_epochs": 11,
    "temperature": 11
  }
}