Hugging Face H4

company

https://github.com/huggingface/alignment-handbook

AI & ML interests

Aligning LLMs to be helpful, honest, harmless, and huggy (H4)

Recent Activity

thomwolf authored a paper 29 days ago

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

lvwerra authored a paper 29 days ago

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

hynky authored a paper 29 days ago

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

View all activity

HuggingFaceH4 's Spaces 18

Zephyr Chat

Chat with an AI model

StarChat2 Demo

Zephyr Gemma Chat

StarChat Playground

Generate coding assistance and answers

Human & GPT-4 Evaluation of LLMs Leaderboard

Falcon-Chat

Interact with Falcon-Chat for personalized conversations

Scaling test-time compute

Implement test-time compute scaling for math problems

Compare Idefics-8b-dpo

Idefics2 8b Playground

Generate text based on images and prompts

Idefics2 8b VDPOed Playground

Running on Zero

VLM Playground

Constitutional AI Demo

Reward Modeling UI

Falcon-Chat (demo for blog post)

Falcon Vs LLaMA

Chatty Lms

Elo Rating

Instruction Model Outputs Filtered