Zobrazeno 1 - 1
of 1
pro vyhledávání: '"Arockiaraj, Benedict Florance"'
We investigate the phenomenon of an LLM's untruthful response using a large set of 220 handcrafted linguistic features. We focus on GPT-3 models and find that the linguistic profiles of responses are similar across model sizes. That is, how varying-s
Externí odkaz:
http://arxiv.org/abs/2305.15875
