lucidrains / memory-efficient-attention-pytorch
Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"
☆360Updated last year
Related projects ⓘ
Alternatives and complementary repositories for memory-efficient-attention-pytorch
- Implementation of fused cosine similarity attention in the same style as Flash Attention☆207Updated last year
- Implementation of Rotary Embeddings, from the Roformer paper, in Pytorch☆571Updated last week
- Memory Efficient Attention (O(sqrt(n)) for Jax and PyTorch☆179Updated last year
- Implementation of Mega, the Single-head Attention with Multi-headed EMA architecture that currently holds SOTA on Long Range Arena☆203Updated last year
- An implementation of local windowed attention for language modeling☆384Updated 2 months ago
- Implementation of the conditionally routed attention in the CoLT5 architecture, in Pytorch☆225Updated 2 months ago
- Implementation of the Adan (ADAptive Nesterov momentum algorithm) Optimizer in Pytorch☆247Updated 2 years ago
- Implementation of Flash Attention in Jax☆196Updated 8 months ago
- Implementation of a Transformer, but completely in Triton☆249Updated 2 years ago
- Helpful tools and examples for working with flex-attention☆469Updated 3 weeks ago
- Implementation of ST-Moe, the latest incarnation of MoE after years of research at Brain, in Pytorch☆293Updated 5 months ago
- A simple way to keep track of an Exponential Moving Average (EMA) version of your Pytorch model☆517Updated last month
- Sequence modeling with Mega.☆298Updated last year
- Official PyTorch Implementation of Long-Short Transformer (NeurIPS 2021).☆222Updated 2 years ago
- Annotated version of the Mamba paper☆457Updated 8 months ago
- FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores☆281Updated last month
- Code for the ALiBi method for transformer language models (ICLR 2022)☆507Updated last year
- Implementation of Soft MoE, proposed by Brain's Vision team, in Pytorch☆246Updated 6 months ago
- Named tensors with first-class dimensions for PyTorch☆322Updated last year
- Implementation of Recurrent Interface Network (RIN), for highly efficient generation of images and video without cascading networks, in P…☆194Updated 9 months ago
- Code release for "Dropout Reduces Underfitting"☆312Updated last year
- Implementation of Block Recurrent Transformer - Pytorch☆213Updated 3 months ago
- Implementation of Linformer for Pytorch☆257Updated 10 months ago
- Transformer based on a variant of attention that is linear complexity in respect to sequence length☆698Updated 6 months ago
- A library that contains a rich collection of performant PyTorch model metrics, a simple interface to create new metrics, a toolkit to fac…☆216Updated last week
- ☆283Updated 2 months ago
- Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT☆205Updated 3 months ago
- Neighborhood Attention Extension. Bringing attention to a neighborhood near you!☆366Updated this week
- ☆132Updated last year