tspeterkim / mixed-precision-from-scratch
Mixed precision training from scratch with Tensors and CUDA
☆18Updated 4 months ago
Related projects: ⓘ
- ☆124Updated 7 months ago
- Cataloging released Triton kernels.☆111Updated 3 weeks ago
- ring-attention experiments☆89Updated 5 months ago
- Small scale distributed training of sequential deep learning models, built on Numpy and MPI.☆88Updated 11 months ago
- Simple and fast low-bit matmul kernels in CUDA☆48Updated this week
- Learn CUDA with PyTorch☆11Updated last month
- An implementation of the transformer architecture onto an Nvidia CUDA kernel☆152Updated 11 months ago
- ☆124Updated last week
- ☆68Updated 2 months ago
- Solve puzzles. Learn CUDA.☆53Updated 9 months ago
- Fast Matrix Multiplications for Lookup Table-Quantized LLMs☆156Updated this week
- A safetensors extension to efficiently store sparse quantized tensors on disk☆26Updated this week
- This repository contains the experimental PyTorch native float8 training UX☆210Updated last month
- ☆66Updated 3 months ago
- ML/DL Math and Method notes☆56Updated 9 months ago
- ☆82Updated 6 months ago
- Applied AI experiments and examples for PyTorch☆123Updated last month
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆66Updated 3 months ago
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆150Updated last week
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆94Updated 2 weeks ago
- Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*☆77Updated 9 months ago
- Experiment of using Tangent to autodiff triton☆66Updated 7 months ago
- This code repository contains the code used for my "Optimizing Memory Usage for Training LLMs and Vision Transformers in PyTorch" blog po…☆84Updated last year
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆73Updated last month
- ☆117Updated 7 months ago
- Collection of kernels written in Triton language☆48Updated 2 weeks ago
- Collection of autoregressive model implementation☆62Updated 2 weeks ago
- Google TPU optimizations for transformers models☆62Updated this week
- ☆13Updated last year
- Gpu benchmark☆35Updated 2 weeks ago