run-28/checkpoint-320/trainer_state.json

{
  "best_metric": 0.6666666666666666,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-28/checkpoint-160",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9870704412460327,
      "learning_rate": 0.0006508102674737513,
      "loss": 0.5724,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.5527419447898865,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8742,
      "eval_samples_per_second": 680.285,
      "eval_steps_per_second": 21.342,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1462595462799072,
      "learning_rate": 0.0005784980155322234,
      "loss": 0.5452,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6533333333333333,
      "eval_f1": 0.42297650130548303,
      "eval_loss": 0.539912760257721,
      "eval_mcc": 0.18167801526026356,
      "eval_precision": 0.4750733137829912,
      "eval_recall": 0.3811764705882353,
      "eval_runtime": 2.2449,
      "eval_samples_per_second": 567.949,
      "eval_steps_per_second": 17.818,
      "step": 320
    }
  ],
  "logging_steps": 500,
  "max_steps": 1600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 583510875840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.891982675867722,
    "learning_rate": 0.0007231225194152792,
    "num_train_epochs": 10,
    "temperature": 2
  }
}