HeegyuKim / torch-xla-SPMDLinks
Pytorch/XLA SPMD Test code in Google TPU
☆23Updated last year
Alternatives and similar repositories for torch-xla-SPMD
Users that are interested in torch-xla-SPMD are comparing it to the libraries listed below
Sorting:
- ☆109Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆80Updated 3 years ago
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆59Updated 8 months ago
- ☆78Updated 11 months ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆70Updated last week
- ☆68Updated 11 months ago
- Official code release for "SuperBPE: Space Travel for Language Models"☆54Updated 2 weeks ago
- Fast, Modern, and Low Precision PyTorch Optimizers☆94Updated this week
- This repo is based on https://github.com/jiaweizzhao/GaLore☆28Updated 9 months ago
- Prune transformer layers☆69Updated last year
- ☆20Updated last year
- Language models scale reliably with over-training and on downstream tasks☆97Updated last year
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆54Updated last year
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆68Updated 10 months ago
- Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*☆82Updated last year
- A fast implementation of T5/UL2 in PyTorch using Flash Attention☆105Updated 3 months ago
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated 2 years ago
- Code for Zero-Shot Tokenizer Transfer☆131Updated 5 months ago
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆80Updated 10 months ago
- ☆126Updated last year
- Experiment of using Tangent to autodiff triton☆79Updated last year
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆127Updated last year
- Truly flash T5 realization!☆67Updated last year
- Simple implementation of Speculative Sampling in NumPy for GPT-2.☆95Updated last year
- ☆49Updated last year
- Experiments for efforts to train a new and improved t5☆77Updated last year
- Implementation of Infini-Transformer in Pytorch☆111Updated 5 months ago
- ☆31Updated last year
- supporting pytorch FSDP for optimizers☆82Updated 6 months ago
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆52Updated 3 months ago