2 ms·
Looks to me like they're using an LLM for _translation_, not for generating a response. The model output even says "Here's the _translation_" (emphasis mine).
by phireal 8mo ago
Looks to me like they're using an LLM for _translation_, not for generating a response. The model output even says "Here's the _translation_" (emphasis mine).