MVP demo of multilingual LLM performance eval space
neutral sd gradio dev space
be polite and rude to llama
generates linkedin posts from freetext entries
compare different llama versions for knowledge cutoff
severely limited context window proof of concept
llm calculator with llama backend
try and get llama to talk about milk
compare different models and their moral compass
simulates the RLHF training step of an LLM
measures CO2e generated from a single query to an LLM
