run-13/checkpoint-428/trainer_state.json

{
  "best_metric": 0.7393087287639133,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-428",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 428,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 5.252918720245361,
      "learning_rate": 0.00015581725710675,
      "loss": 0.4793,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6994727592267135,
      "eval_f1": 0.2982216142270862,
      "eval_loss": 0.45286598801612854,
      "eval_mcc": 0.23321001518594284,
      "eval_precision": 0.6728395061728395,
      "eval_recall": 0.19156414762741653,
      "eval_runtime": 3.1418,
      "eval_samples_per_second": 543.327,
      "eval_steps_per_second": 17.188,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6013866662979126,
      "learning_rate": 0.00013850422853933333,
      "loss": 0.4377,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7393087287639133,
      "eval_f1": 0.44165621079046424,
      "eval_loss": 0.4278346300125122,
      "eval_mcc": 0.36530426808949645,
      "eval_precision": 0.7719298245614035,
      "eval_recall": 0.3093145869947276,
      "eval_runtime": 3.129,
      "eval_samples_per_second": 545.542,
      "eval_steps_per_second": 17.258,
      "step": 428
    }
  ],
  "logging_steps": 500,
  "max_steps": 2140,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1049551328880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7600265334081553,
    "learning_rate": 0.00017313028567416666,
    "num_train_epochs": 10,
    "temperature": 20
  }
}