Adaptive Pre-training Data Detection for Large Language Models via Surprising Tokens

Devs

Adaptive Pre-training Data Detection for Large Language Models via Surprising Tokens | Read Paper on Bytez