glassroom / heinsen_attentionLinks
Reference implementation of "Softmax Attention with Constant Cost per Token" (Heinsen, 2024)
☆24Updated 11 months ago
Alternatives and similar repositories for heinsen_attention
Users that are interested in heinsen_attention are comparing it to the libraries listed below
Sorting:
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- ☆31Updated last year
- ☆46Updated last year
- HGRN2: Gated Linear RNNs with State Expansion☆54Updated 9 months ago
- ☆11Updated last year
- ☆23Updated 8 months ago
- Official Code Repository for the paper "Key-value memory in the brain"☆26Updated 3 months ago
- Here we will test various linear attention designs.☆58Updated last year
- sigma-MoE layer☆18Updated last year
- ☆20Updated last year
- ☆31Updated last year
- Official code for the paper "Attention as a Hypernetwork"☆36Updated 11 months ago
- ☆54Updated 10 months ago
- Efficient PScan implementation in PyTorch☆16Updated last year
- Combining SOAP and MUON☆16Updated 3 months ago
- Xmixers: A collection of SOTA efficient token/channel mixers☆11Updated 6 months ago
- Official repository for the paper "Approximating Two-Layer Feedforward Networks for Efficient Transformers"☆37Updated last year
- Code for the paper: https://arxiv.org/pdf/2309.06979.pdf☆19Updated 10 months ago
- ☆15Updated last year
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated last year
- Using FlexAttention to compute attention with different masking patterns☆43Updated 8 months ago
- Triton implement of bi-directional (non-causal) linear attention☆48Updated 4 months ago
- This repository contains code for the MicroAdam paper.☆18Updated 5 months ago
- [NeurIPS 2023 spotlight] Official implementation of HGRN in our NeurIPS 2023 paper - Hierarchically Gated Recurrent Neural Network for Se…☆64Updated last year
- 32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.☆48Updated last year
- Blog post☆17Updated last year
- ☆21Updated last year
- Experiments on the impact of depth in transformers and SSMs.☆30Updated 6 months ago
- User-friendly implementation of the Mixture-of-Sparse-Attention (MoSA). MoSA selects distinct tokens for each head with expert choice rou …☆16Updated last month
- APPy (Annotated Parallelism for Python) enables users to annotate loops and tensor expressions in Python with compiler directives akin to…☆23Updated 2 weeks ago