run-6/checkpoint-1920/trainer_state.json

{
  "best_metric": 0.8903107861060329,
  "best_model_checkpoint": "distilbert-hate_speech18\\run-6\\checkpoint-1680",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 1920,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.4189080485362326e-06,
      "loss": 0.4732,
      "step": 240
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8706581352833638,
      "eval_loss": 0.45438072085380554,
      "eval_runtime": 8.0476,
      "eval_samples_per_second": 271.882,
      "eval_steps_per_second": 17.024,
      "step": 240
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.7876354701739137e-06,
      "loss": 0.4459,
      "step": 480
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8857404021937842,
      "eval_loss": 0.4488455057144165,
      "eval_runtime": 7.9963,
      "eval_samples_per_second": 273.628,
      "eval_steps_per_second": 17.133,
      "step": 480
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.1563628918115945e-06,
      "loss": 0.4356,
      "step": 720
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8893967093235832,
      "eval_loss": 0.4489736258983612,
      "eval_runtime": 8.1712,
      "eval_samples_per_second": 267.77,
      "eval_steps_per_second": 16.766,
      "step": 720
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.5250903134492757e-06,
      "loss": 0.4283,
      "step": 960
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8889396709323584,
      "eval_loss": 0.44683799147605896,
      "eval_runtime": 8.1009,
      "eval_samples_per_second": 270.094,
      "eval_steps_per_second": 16.912,
      "step": 960
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.8938177350869569e-06,
      "loss": 0.4229,
      "step": 1200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8889396709323584,
      "eval_loss": 0.4480336308479309,
      "eval_runtime": 8.1801,
      "eval_samples_per_second": 267.477,
      "eval_steps_per_second": 16.748,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.2625451567246378e-06,
      "loss": 0.4196,
      "step": 1440
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8871115173674589,
      "eval_loss": 0.450763076543808,
      "eval_runtime": 8.0343,
      "eval_samples_per_second": 272.332,
      "eval_steps_per_second": 17.052,
      "step": 1440
    },
    {
      "epoch": 7.0,
      "learning_rate": 6.312725783623189e-07,
      "loss": 0.4169,
      "step": 1680
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8903107861060329,
      "eval_loss": 0.4496695101261139,
      "eval_runtime": 8.0794,
      "eval_samples_per_second": 270.811,
      "eval_steps_per_second": 16.957,
      "step": 1680
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0,
      "loss": 0.4159,
      "step": 1920
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8898537477148081,
      "eval_loss": 0.4497666656970978,
      "eval_runtime": 8.166,
      "eval_samples_per_second": 267.941,
      "eval_steps_per_second": 16.777,
      "step": 1920
    }
  ],
  "logging_steps": 500,
  "max_steps": 1920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1306917187523808.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4787113643124785,
    "learning_rate": 5.050180626898551e-06,
    "num_train_epochs": 8,
    "per_device_eval_batch_size": 16,
    "per_device_train_batch_size": 32,
    "temperature": 27
  }
}