↓ 2 callersFunctionppo_train(ref_logits: list[float], reward: dict[str, float],
beta: float, steps: int = 300, batch: int =
phases/18-ethics-safety-alignment/04-sycophancy-rlhf-amplification/code/main.py:71
↓ 2 callersMethodrecord(self, user_id, model, input_tokens, output_tokens, cost)
phases/11-llm-engineering/13-production-app/code/production_app.py:59