Uncensor any LLM with abliteration

Not All Language Model Features Are Linear

ConvLLaVA: Hierarchical Backbones as Visual Encoder for Large Multimodal Models

LLM: Transformer Is Linear

Idefics2: A Powerful 8B Vision-Language Model for the Community

Llava Model for Video

Mini- Dust3r: A miniature version of dust3r running in a HuggingFace Space

An Introduction to Vision-Language Modeling

ML for 3D Course on Hugging Face

Llama-3-70B-Instruct-Gradient-1048k