{
  "best_metric": 0.7586408904510837,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-23/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": 4.89791202545166,
      "learning_rate": 0.0003087091816763773,
      "loss": 0.4811,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7234915055653193,
      "eval_f1": 0.42857142857142855,
      "eval_loss": 0.44225630164146423,
      "eval_mcc": 0.3173839802242578,
      "eval_precision": 0.688715953307393,
      "eval_recall": 0.3110720562390158,
      "eval_runtime": 3.1485,
      "eval_samples_per_second": 542.164,
      "eval_steps_per_second": 17.151,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.2303036451339722,
      "learning_rate": 0.00028298341653667917,
      "loss": 0.4399,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7293497363796133,
      "eval_f1": 0.384,
      "eval_loss": 0.43119725584983826,
      "eval_mcc": 0.3377086381007214,
      "eval_precision": 0.7955801104972375,
      "eval_recall": 0.2530755711775044,
      "eval_runtime": 3.2127,
      "eval_samples_per_second": 531.328,
      "eval_steps_per_second": 16.808,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7242010831832886,
      "learning_rate": 0.0002572576513969811,
      "loss": 0.4216,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7439953134153485,
      "eval_f1": 0.450314465408805,
      "eval_loss": 0.42299720644950867,
      "eval_mcc": 0.38011380476662987,
      "eval_precision": 0.7920353982300885,
      "eval_recall": 0.3145869947275923,
      "eval_runtime": 3.144,
      "eval_samples_per_second": 542.939,
      "eval_steps_per_second": 17.176,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.6966323852539062,
      "learning_rate": 0.00023153188625728296,
      "loss": 0.4166,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7451669595782073,
      "eval_f1": 0.45692883895131087,
      "eval_loss": 0.4213677942752838,
      "eval_mcc": 0.3831812605460561,
      "eval_precision": 0.7887931034482759,
      "eval_recall": 0.3216168717047452,
      "eval_runtime": 3.2205,
      "eval_samples_per_second": 530.042,
      "eval_steps_per_second": 16.768,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.2818307876586914,
      "learning_rate": 0.00020580612111758488,
      "loss": 0.408,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7586408904510837,
      "eval_f1": 0.5502183406113537,
      "eval_loss": 0.41744720935821533,
      "eval_mcc": 0.42099226702686965,
      "eval_precision": 0.7262247838616714,
      "eval_recall": 0.4428822495606327,
      "eval_runtime": 3.155,
      "eval_samples_per_second": 541.043,
      "eval_steps_per_second": 17.116,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 2782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 13,
  "save_steps": 500,
  "total_flos": 2623878322200.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7693731022825507,
    "learning_rate": 0.0003344349468160754,
    "num_train_epochs": 13,
    "temperature": 28
  }
}