Hacker News new | ask | show | jobs
by errantspark 631 days ago
The claim is that llama is "lobotomized" because it was trained with safety in mind. You can't untrain that by finetuning. For what it's worth the non-instruct llama generally seems better at reasoning than instruct llama which i think is a point in support of OP.
1 comments

Better at reasoning based on benchmarks or what?