@voooooogel 2023-03-20 ♥3 ↻0 original ↗
@reconfigurthing @elymitra_ personally I've tried llama 13B (quantized via llama.cpp tbf) and it really didn't feel GPT-3.5-tier to me.

granted i haven't tried any of the alpaca RL*F'd versions

author:voooooogel kind:tweet model:gpt-3-5 on:llama-1 year:2023

cited on: llama-1

Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.