Lewis Tunstall's picture

Lewis Tunstall PRO

lewtun

·

https://lewtun.github.io/blog/

AI & ML interests

LLMs, LLMs, LLMs

Recent Activity

liked a Space 1 day ago

HuggingFaceH4/blogpost-scaling-test-time-compute

posted an update 1 day ago

We outperform Llama 70B with Llama 3B on hard math by scaling test-time compute 🔥 How? By combining step-wise reward models with tree search algorithms :) We show that smol models can match or exceed the performance of their much larger siblings when given enough "time to think" We're open sourcing the full recipe and sharing a detailed blog post. In our blog post we cover: 📈 Compute-optimal scaling: How we implemented DeepMind's recipe to boost the mathematical capabilities of open models at test-time. 🎄 Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets. 🧭 Search and Learn: A lightweight toolkit for implementing search strategies with LLMs and built for speed with vLLM Here's the links: - Blog post: https://huggingface.co/spaces/HuggingFaceH4/blogpost-scaling-test-time-compute - Code: https://github.com/huggingface/search-and-learn Enjoy!

New activity 1 day ago

HuggingFaceH4/blogpost-scaling-test-time-compute:Use HF papers

View all activity

Articles

Universal Assisted Generation: Faster Decoding with Any Assistant Model

Faster Assisted Generation with Dynamic Speculation

Llama can now see and run on your device - welcome Llama 3.2

FineVideo: behind the scenes

How NuminaMath Won the 1st AIMO Progress Prize

Welcome Gemma 2 - Google's new open LLM

Constitutional AI with Open LLMs

Preference Tuning LLMs with Direct Preference Optimization Methods

Mixture of Experts Explained

Welcome Mixtral - a SOTA Mixture of Experts on Hugging Face

SetFitABSA: Few-Shot Aspect Based Sentiment Analysis using SetFit

Fine-tuning Llama 2 70B using PyTorch FSDP

Code Llama: Llama 2 learns to code

Llama 2 is here - get it on Hugging Face

Can foundation models label data like humans?

The Falcon has landed in the Hugging Face ecosystem

Creating a Coding Assistant with StarCoder

StackLLaMA: A hands-on guide to train LLaMA with RLHF

Fine-tuning 20B LLMs with RLHF on a 24GB consumer GPU

Red-Teaming Large Language Models

Diffusion Models Live Event

Very Large Language Models and How to Evaluate Them

SetFit: Efficient Few-Shot Learning Without Prompts

Announcing Evaluation on the Hub

Organizations

lewtun's activity

authored 3 papers 12 months ago

RAFT: A Real-World Few-Shot Text Classification Benchmark

Paper • 2109.14076 • Published Sep 28, 2021 • 2

Efficient Few-Shot Learning Without Prompts

Paper • 2209.11055 • Published Sep 22, 2022 • 3

GEMv2: Multilingual NLG Benchmarking in a Single Line of Code

Paper • 2206.11249 • Published Jun 22, 2022

authored a paper about 1 year ago

Zephyr: Direct Distillation of LM Alignment

Paper • 2310.16944 • Published Oct 25, 2023 • 122

authored a paper over 1 year ago

AfroDigits: A Community-Driven Spoken Digit Dataset for African Languages

Paper • 2303.12582 • Published Mar 22, 2023 • 20

authored a paper almost 2 years ago

Evaluate & Evaluation on the Hub: Better Best Practices for Data and Model Measurements

Paper • 2210.01970 • Published Sep 30, 2022 • 11