For the traditional benchmarks on OpenLLM Leaderboard, most of the drop happens in TruthfulQA and GSM8k. However, in return you get impressive performance on RULER. gradient.ai/blog/evaluating-…

@Gradient_AI_ owes a lot to @huggingface and @ClemDelangue for supporting the community. Honored to meet you. Also thanks to our investor @Wing_VC for hosting. (I’ll make sure to keep my eyes open next time 🤣)

From the OpenLLM Leaderboard. We're working on extensive evaluations.

Number 5 and Number 8 on @huggingface More improvements coming out...

As a subscriber to AINews it's cool to see our model getting picked up in the summarizer. My fav daily recap @swyx

3 things I love about this image: 1. Lots of Llamas 2. We got the longest Llama in the top 20 3. Also seeing others cook 🧑‍🍳 @mattshumer_ source go get it @ClementDelangue more Long Llamas going up Hub!

I’m excited to be talking with @dpbrinkm and Ben Epstein on May 17th, Wednesday for the @mlopscommunity on Twitter meetup about Fine-Tuning LLMs: Best Practices and When to Go Small. Register to join us here: home.mlops.community/home/ev…