llmsearch

chain-of-hindsight

Chain-of-Hindsight, A Scalable RLHF Method

starcoder2

Home of StarCoder2!

digital_palace

My Digital Palace - A Personal Journal for Reflection - A place to store all my thoughts

TUPE

Transformer with Untied Positional Encoding (TUPE). Code of paper "Rethinking Positional Encoding...

plain-transformers

Transformer models implementation for training from scratch.

sentiment-discovery

Unsupervised Language Modeling at scale for robust sentiment classification

ChatAbstractions

LangChain chat model abstractions for dynamic failover, load balancing, chaos engineering, and more!

multi_token

Embed arbitrary modalities (images, audio, documents, etc) into large language models.

self-rag

This includes the original implementation of SELF-RAG: Learning to Retrieve, Generate and Critiqu...

long_llama

LongLLaMA is a large language model capable of handling long contexts. It is based on OpenLLaMA a...

Megatron-LM

Ongoing research training transformer models at scale

hybrid-discriminative-generative

Hybrid Discriminative-Generative Training via Contrastive Learning

accelerated-pytorch-transformers-generation

textbook_quality

Generate textbook-quality synthetic LLM pretraining data

starcoder

Home of StarCoder: fine-tuning & inference!