glassroom / heinsen_attentionLinks
Reference implementation of "Softmax Attention with Constant Cost per Token" (Heinsen, 2024)
☆24Updated last year
Alternatives and similar repositories for heinsen_attention
Users that are interested in heinsen_attention are comparing it to the libraries listed below
Sorting:
- Here we will test various linear attention designs.☆62Updated last year
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- ☆49Updated last year
- ☆23Updated 10 months ago
- ☆32Updated last year
- [NeurIPS 2023 spotlight] Official implementation of HGRN in our NeurIPS 2023 paper - Hierarchically Gated Recurrent Neural Network for Se…☆66Updated last year
- ☆20Updated last year
- ☆11Updated last year
- ☆56Updated last year
- Official Code Repository for the paper "Key-value memory in the brain"☆27Updated 5 months ago
- Expanding linear RNN state-transition matrix eigenvalues to include negatives improves state-tracking tasks and language modeling without…☆15Updated 4 months ago
- ☆33Updated last year
- Awesome Triton Resources☆32Updated 3 months ago
- Xmixers: A collection of SOTA efficient token/channel mixers☆11Updated 3 weeks ago
- Experiments on the impact of depth in transformers and SSMs.☆32Updated 8 months ago
- Efficient PScan implementation in PyTorch☆16Updated last year
- RADLADS training code☆26Updated 2 months ago
- Official code for the paper "Attention as a Hypernetwork"☆40Updated last year
- HGRN2: Gated Linear RNNs with State Expansion☆55Updated 11 months ago
- Combining SOAP and MUON☆16Updated 5 months ago
- Implementation of Hyena Hierarchy in JAX☆10Updated 2 years ago
- Triton implement of bi-directional (non-causal) linear attention☆52Updated 6 months ago
- sigma-MoE layer☆20Updated last year
- [ICLR 2025] Official PyTorch implementation of "Forgetting Transformer: Softmax Attention with a Forget Gate"☆118Updated last month
- Stick-breaking attention☆58Updated last month
- Linear Attention Sequence Parallelism (LASP)☆85Updated last year
- Griffin MQA + Hawk Linear RNN Hybrid☆88Updated last year
- [NeurIPS 2023] Sparse Modular Activation for Efficient Sequence Modeling☆37Updated last year
- APPy (Annotated Parallelism for Python) enables users to annotate loops and tensor expressions in Python with compiler directives akin to…☆24Updated last month
- Triton Implementation of HyperAttention Algorithm☆48Updated last year