{
  "best_metric": 0.7615686274509804,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-12/checkpoint-1120",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 1280,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.5787363052368164,
      "learning_rate": 0.00024237605415726302,
      "loss": 0.4097,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6658823529411765,
      "eval_f1": 0.009302325581395347,
      "eval_loss": 0.38458171486854553,
      "eval_mcc": 0.008873565094161137,
      "eval_precision": 0.4,
      "eval_recall": 0.004705882352941176,
      "eval_runtime": 1.8643,
      "eval_samples_per_second": 683.908,
      "eval_steps_per_second": 21.456,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8726533055305481,
      "learning_rate": 0.0002077509035633683,
      "loss": 0.3859,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7027450980392157,
      "eval_f1": 0.2612085769980507,
      "eval_loss": 0.37138044834136963,
      "eval_mcc": 0.24722748455315502,
      "eval_precision": 0.7613636363636364,
      "eval_recall": 0.15764705882352942,
      "eval_runtime": 1.863,
      "eval_samples_per_second": 684.383,
      "eval_steps_per_second": 21.471,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1823008060455322,
      "learning_rate": 0.00017312575296947358,
      "loss": 0.3761,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7419607843137255,
      "eval_f1": 0.5168869309838473,
      "eval_loss": 0.3662872314453125,
      "eval_mcc": 0.3765709652306714,
      "eval_precision": 0.6875,
      "eval_recall": 0.41411764705882353,
      "eval_runtime": 1.8612,
      "eval_samples_per_second": 685.044,
      "eval_steps_per_second": 21.492,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8506985306739807,
      "learning_rate": 0.00013850060237557887,
      "loss": 0.3709,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7231372549019608,
      "eval_f1": 0.3796133567662565,
      "eval_loss": 0.36449602246284485,
      "eval_mcc": 0.315387850148385,
      "eval_precision": 0.75,
      "eval_recall": 0.2541176470588235,
      "eval_runtime": 1.8672,
      "eval_samples_per_second": 682.835,
      "eval_steps_per_second": 21.422,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.953569233417511,
      "learning_rate": 0.00010387545178168416,
      "loss": 0.3673,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7537254901960785,
      "eval_f1": 0.5552407932011331,
      "eval_loss": 0.3644358813762665,
      "eval_mcc": 0.41074948302085584,
      "eval_precision": 0.697508896797153,
      "eval_recall": 0.4611764705882353,
      "eval_runtime": 1.8615,
      "eval_samples_per_second": 684.938,
      "eval_steps_per_second": 21.488,
      "step": 800
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.1368662118911743,
      "learning_rate": 6.925030118778943e-05,
      "loss": 0.3635,
      "step": 960
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7482352941176471,
      "eval_f1": 0.5172932330827068,
      "eval_loss": 0.3630056381225586,
      "eval_mcc": 0.3915780041490244,
      "eval_precision": 0.7166666666666667,
      "eval_recall": 0.4047058823529412,
      "eval_runtime": 1.8707,
      "eval_samples_per_second": 681.562,
      "eval_steps_per_second": 21.382,
      "step": 960
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.311170220375061,
      "learning_rate": 3.4625150593894717e-05,
      "loss": 0.363,
      "step": 1120
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7615686274509804,
      "eval_f1": 0.5694050991501417,
      "eval_loss": 0.3646318018436432,
      "eval_mcc": 0.43081867600233087,
      "eval_precision": 0.7153024911032029,
      "eval_recall": 0.47294117647058825,
      "eval_runtime": 1.8727,
      "eval_samples_per_second": 680.818,
      "eval_steps_per_second": 21.359,
      "step": 1120
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.0805109739303589,
      "learning_rate": 0.0,
      "loss": 0.3592,
      "step": 1280
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7513725490196078,
      "eval_f1": 0.5261584454409567,
      "eval_loss": 0.3618643283843994,
      "eval_mcc": 0.40038619635473977,
      "eval_precision": 0.7213114754098361,
      "eval_recall": 0.41411764705882353,
      "eval_runtime": 1.8765,
      "eval_samples_per_second": 679.461,
      "eval_steps_per_second": 21.316,
      "step": 1280
    }
  ],
  "logging_steps": 500,
  "max_steps": 1280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 2334043503360.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6289598397405775,
    "learning_rate": 0.00027700120475115773,
    "num_train_epochs": 8,
    "temperature": 22
  }
}