salesforce / jaxformer
Minimal library to train LLMs on TPU in JAX with pjit().
☆283Updated last year
Alternatives and similar repositories for jaxformer:
Users that are interested in jaxformer are comparing it to the libraries listed below
- Repository for analysis and experiments in the BigCode project.☆117Updated last year
- Generative model for code infilling and synthesis☆300Updated last year
- CodeGen2 models for program synthesis☆274Updated last year
- Code for the paper "Efficient Training of Language Models to Fill in the Middle"☆175Updated last year
- This is the official code for the paper CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning (Neur…☆521Updated 2 months ago
- Ongoing research training transformer models at scale☆383Updated 7 months ago
- GPTQ inference Triton kernel☆298Updated last year
- Open Instruction Generalist is an assistant trained on massive synthetic instructions to perform many millions of tasks☆208Updated last year
- Fast Inference Solutions for BLOOM☆562Updated 5 months ago
- ☆268Updated last year
- Fine-tune SantaCoder for Code/Text Generation.☆190Updated last year
- Train very large language models in Jax.☆203Updated last year
- Accepted by Transactions on Machine Learning Research (TMLR)☆126Updated 5 months ago
- ☆423Updated 7 months ago
- batched loras☆340Updated last year
- Code for the paper "Rethinking Benchmark and Contamination for Language Models with Rephrased Samples"☆298Updated last year
- Official repository for LongChat and LongEval☆516Updated 10 months ago
- ☆542Updated 3 months ago
- Landmark Attention: Random-Access Infinite Context Length for Transformers☆421Updated last year
- A framework for the evaluation of autoregressive code generation language models.☆912Updated 4 months ago
- ☆51Updated 2 weeks ago
- 🐙 OctoPack: Instruction Tuning Code Large Language Models☆459Updated last month
- [NeurIPS 2023 D&B] Code repository for InterCode benchmark https://arxiv.org/abs/2306.14898☆210Updated 10 months ago
- Inference code for LLaMA models in JAX☆116Updated 10 months ago
- ☆355Updated 2 years ago
- The data processing pipeline for the Koala chatbot language model☆117Updated last year
- Salesforce open-source LLMs with 8k sequence length.☆716Updated last month
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆217Updated 11 months ago
- Multipack distributed sampler for fast padding-free training of LLMs☆186Updated 7 months ago
- An open collection of methodologies to help with successful training of large language models.☆480Updated last year