Building modular LMs with parameter-efficient fine-tuning.
MIT License
Bot releases are hidden (Show)
Published by matheper 8 months ago
This code release accompanies the paper Multi-Head Adapter Routing for Cross-Task Generalization
Shared Middle-Layer for Triton Compilation
AICI: Prompts as (Wasm) Programs
Community for applying LLMs to robotics and a robot simulator with ChatGPT integration
MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.
Trace, the New AutoDiff for AI Systems and LLM Agents
Repo for WWW 2022 paper: Progressively Optimized Bi-Granular Document Representation for Scalable...
Official Codebase for MEGAVERSE: (published in ACL: NAACL 2024)
MSCCL++: A GPU-driven communication stack for scalable AI applications
To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention,...
Tutel MoE: An Optimized Mixture-of-Experts Implementation
Library to convert natural language utterance into a structured domain specific language