fattorib / ZeRO-transformerLinks
Two implementations of ZeRO-1 optimizer sharding in JAX
☆14Updated 2 years ago
Alternatives and similar repositories for ZeRO-transformer
Users that are interested in ZeRO-transformer are comparing it to the libraries listed below
Sorting:
- seqax = sequence modeling + JAX☆168Updated 3 months ago
- Minimal but scalable implementation of large language models in JAX☆35Updated last month
- ☆103Updated this week
- JAX bindings for Flash Attention v2☆97Updated last week
- Distributed pretraining of large language models (LLMs) on cloud TPU slices, with Jax and Equinox.☆24Updated last year
- ☆91Updated last year
- Minimal yet performant LLM examples in pure JAX☆187Updated last month
- A simple library for scaling up JAX programs☆144Updated 11 months ago
- Experiment of using Tangent to autodiff triton☆80Updated last year
- A library for unit scaling in PyTorch☆132Updated 3 months ago
- ☆283Updated last year
- Inference code for LLaMA models in JAX☆119Updated last year
- ☆190Updated this week
- If it quacks like a tensor...☆59Updated 11 months ago
- An implementation of the Llama architecture, to instruct and delight☆21Updated 4 months ago
- LoRA for arbitrary JAX models and functions☆141Updated last year
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆45Updated last year
- JAX implementation of the Llama 2 model☆216Updated last year
- Train very large language models in Jax.☆209Updated 2 years ago
- ☆62Updated 3 years ago
- JAX implementation of the Mistral 7b v0.2 model☆35Updated last year
- Machine Learning eXperiment Utilities☆46Updated 3 months ago
- A set of Python scripts that makes your experience on TPU better☆54Updated last month
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- Solve puzzles. Learn CUDA.☆64Updated last year
- Accelerated First Order Parallel Associative Scan☆189Updated last year
- Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.☆32Updated 4 months ago
- A flexible and efficient implementation of Flash Attention 2.0 for JAX, supporting multiple backends (GPU/TPU/CPU) and platforms (Triton/…☆28Updated 7 months ago
- supporting pytorch FSDP for optimizers☆83Updated 10 months ago
- ☆335Updated last month