salesforce / jaxformerLinks
Minimal library to train LLMs on TPU in JAX with pjit().
☆299Updated last year
Alternatives and similar repositories for jaxformer
Users that are interested in jaxformer are comparing it to the libraries listed below
Sorting:
- Repository for analysis and experiments in the BigCode project.☆128Updated last year
- DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.☆171Updated 2 months ago
- Code for the paper "Efficient Training of Language Models to Fill in the Middle"☆195Updated 2 years ago
- Inference code for LLaMA models in JAX☆120Updated last year
- Fine-tune SantaCoder for Code/Text Generation.☆194Updated 2 years ago
- CodeGen2 models for program synthesis☆271Updated 2 years ago
- Train very large language models in Jax.☆210Updated 2 years ago
- Ongoing research training transformer models at scale☆394Updated last year
- Multipack distributed sampler for fast padding-free training of LLMs☆202Updated last year
- Used for adaptive human in the loop evaluation of language and embedding models.☆308Updated 2 years ago
- Generative model for code infilling and synthesis☆308Updated 2 years ago
- This is the official code for the paper CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning (Neur…☆560Updated 10 months ago
- Accepted by Transactions on Machine Learning Research (TMLR)☆136Updated last year
- NeurIPS Large Language Model Efficiency Challenge: 1 LLM + 1GPU + 1Day☆258Updated 2 years ago
- Code for the paper "Rethinking Benchmark and Contamination for Language Models with Rephrased Samples"☆315Updated last year
- Scaling Data-Constrained Language Models☆342Updated 5 months ago
- The data processing pipeline for the Koala chatbot language model☆118Updated 2 years ago
- JAX implementation of the Llama 2 model☆216Updated last year
- A hard gym for programming☆162Updated last year
- ☆379Updated 2 years ago
- Blazing fast training of 🤗 Transformers on Graphcore IPUs☆86Updated last year
- Minimal code to train a Large Language Model (LLM).☆172Updated 3 years ago
- ☆277Updated 2 years ago
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆226Updated 3 months ago
- This is the repo for the paper Shepherd -- A Critic for Language Model Generation☆220Updated 2 years ago
- Experiments on speculative sampling with Llama models☆127Updated 2 years ago
- Open Instruction Generalist is an assistant trained on massive synthetic instructions to perform many millions of tasks☆210Updated last year
- A repository for research on medium sized language models.☆522Updated 6 months ago
- batched loras☆347Updated 2 years ago
- A crude RLHF layer on top of nanoGPT with Gumbel-Softmax trick☆293Updated 2 years ago