Chat Templates for 🤗 HuggingFace Large Language Models

Together Mixture-Of-Agents (MoA) – 65.1% on AlpacaEval with OSS models

The source code for the blog post The 37 Implementation Details of Proximal Policy Optimization

Recipes to train reward model for RLHF.

Benchmarking LLMs with Challenging Tasks from Real Users

Let ChatGPT teach your own chatbot in hours with a single GPU!

Dromedary: towards helpful, ethical and reliable LLMs.

Large-scale, Informative, and Diverse Multi-round Chat Data (and Models)

Official repository for "Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing". Your efficient and high-quality synthetic data generation pipeline!

Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

Simple Python library/structure to ablate features in LLMs which are supported by TransformerLens

Qwen2 is the large language model series developed by Qwen team, Alibaba Cloud.

A set of scripts to grab public datasets from resources related to arXiv

A curated list of resources dedicated to Python libraries, LLMs, dictionaries, and corpora of NLP for Japanese

RAG AutoML Tool - Find optimal RAG pipeline for your own data.

MoRA: High-Rank Updating for Parameter-Efficient Fine-Tuning

Official repo for "Make Your LLM Fully Utilize the Context"

CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Korean

Virtual whiteboard for sketching hand-drawn like diagrams

llama3 implementation one matrix multiplication at a time

A high-throughput and memory-efficient inference and serving engine for LLMs

Holistic Evaluation of Language Models (HELM), a framework to increase the transparency of language models (https://arxiv.org/abs/2211.09110). This framework is also used to evaluate text-to-image …

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

A very fast and expressive template engine.

Official implementation of ECCV2024 paper, "DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs".