{
  "best_metric": 0.6831683168316832,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-1485",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1782,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.1475601196289062,
      "learning_rate": 0.00023316325483820226,
      "loss": 0.6759,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5623762376237624,
      "eval_f1": 0.3402985074626866,
      "eval_loss": 0.672980785369873,
      "eval_mcc": 0.1665189802603812,
      "eval_precision": 0.6867469879518072,
      "eval_recall": 0.2261904761904762,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 539.729,
      "eval_steps_per_second": 17.1,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.315560817718506,
      "learning_rate": 0.0001865306038705618,
      "loss": 0.6336,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6475247524752475,
      "eval_f1": 0.6941580756013745,
      "eval_loss": 0.623160719871521,
      "eval_mcc": 0.3106524830468984,
      "eval_precision": 0.6121212121212121,
      "eval_recall": 0.8015873015873016,
      "eval_runtime": 0.9321,
      "eval_samples_per_second": 541.793,
      "eval_steps_per_second": 17.166,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.8556504249572754,
      "learning_rate": 0.00013989795290292136,
      "loss": 0.6023,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6336633663366337,
      "eval_f1": 0.7049441786283892,
      "eval_loss": 0.6375722885131836,
      "eval_mcc": 0.30681404579985594,
      "eval_precision": 0.5893333333333334,
      "eval_recall": 0.876984126984127,
      "eval_runtime": 0.9395,
      "eval_samples_per_second": 537.528,
      "eval_steps_per_second": 17.031,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2792253494262695,
      "learning_rate": 9.32653019352809e-05,
      "loss": 0.5844,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6495049504950495,
      "eval_f1": 0.7015177065767284,
      "eval_loss": 0.6306431889533997,
      "eval_mcc": 0.32000413347466017,
      "eval_precision": 0.6099706744868035,
      "eval_recall": 0.8253968253968254,
      "eval_runtime": 0.9346,
      "eval_samples_per_second": 540.359,
      "eval_steps_per_second": 17.12,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.801590919494629,
      "learning_rate": 4.663265096764045e-05,
      "loss": 0.5623,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6831683168316832,
      "eval_f1": 0.7122302158273381,
      "eval_loss": 0.5899890661239624,
      "eval_mcc": 0.3746159847803109,
      "eval_precision": 0.6513157894736842,
      "eval_recall": 0.7857142857142857,
      "eval_runtime": 0.9473,
      "eval_samples_per_second": 533.119,
      "eval_steps_per_second": 16.891,
      "step": 1485
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.276622772216797,
      "learning_rate": 0.0,
      "loss": 0.5457,
      "step": 1782
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6831683168316832,
      "eval_f1": 0.7153024911032029,
      "eval_loss": 0.5890778303146362,
      "eval_mcc": 0.3766853158810184,
      "eval_precision": 0.6483870967741936,
      "eval_recall": 0.7976190476190477,
      "eval_runtime": 0.938,
      "eval_samples_per_second": 538.372,
      "eval_steps_per_second": 17.057,
      "step": 1782
    }
  ],
  "logging_steps": 500,
  "max_steps": 1782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 4384207595880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9735237137483809,
    "learning_rate": 0.0002797959058058427,
    "num_train_epochs": 6,
    "temperature": 49
  }
}