run-1/checkpoint-110/trainer_state.json

{
  "best_metric": 0.3939393939393939,
  "best_model_checkpoint": "tinybert-toxigen-bothpretrained\\run-1\\checkpoint-110",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 110,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 2.266610127439989e-05,
      "loss": 0.1001,
      "step": 55
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.36262626262626263,
      "eval_loss": 0.09902925044298172,
      "eval_runtime": 3.347,
      "eval_samples_per_second": 591.576,
      "eval_steps_per_second": 4.78,
      "step": 55
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.511073418293326e-05,
      "loss": 0.0993,
      "step": 110
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3939393939393939,
      "eval_loss": 0.09844165295362473,
      "eval_runtime": 3.4042,
      "eval_samples_per_second": 581.63,
      "eval_steps_per_second": 4.7,
      "step": 110
    }
  ],
  "logging_steps": 500,
  "max_steps": 220,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1682283651840.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.13686313401622996,
    "learning_rate": 3.022146836586652e-05,
    "num_train_epochs": 4,
    "temperature": 18
  }
}