lucidrains / mixture-of-attentionLinks
Some personal experiments around routing tokens to different autoregressive attention, akin to mixture-of-experts
☆122Updated last year
Alternatives and similar repositories for mixture-of-attention
Users that are interested in mixture-of-attention are comparing it to the libraries listed below
Sorting:
- Implementation of Infini-Transformer in Pytorch☆112Updated last year
- Explorations into the recently proposed Taylor Series Linear Attention☆100Updated last year
- Implementation of GateLoop Transformer in Pytorch and Jax☆92Updated last year
- Exploration into the proposed "Self Reasoning Tokens" by Felipe Bonetto☆57Updated last year
- Implementation of 🌻 Mirasol, SOTA Multimodal Autoregressive model out of Google Deepmind, in Pytorch☆91Updated 2 years ago
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆53Updated 2 years ago
- Implementation of Agent Attention in Pytorch☆93Updated last year
- Exploring an idea where one forgets about efficiency and carries out attention across each edge of the nodes (tokens)☆55Updated 10 months ago
- Implementation of the Kalman Filtering Attention proposed in "Kalman Filtering Attention for User Behavior Modeling in CTR Prediction"☆59Updated 2 years ago
- Implementation of a multimodal diffusion transformer in Pytorch☆107Updated last year
- Implementation of the Llama architecture with RLHF + Q-learning☆170Updated 11 months ago
- PyTorch implementation of Soft MoE by Google Brain in "From Sparse to Soft Mixtures of Experts" (https://arxiv.org/pdf/2308.00951.pdf)☆82Updated 2 years ago
- CUDA implementation of autoregressive linear attention, with all the latest research findings☆46Updated 2 years ago
- Implementation of Gated State Spaces, from the paper "Long Range Language Modeling via Gated State Spaces", in Pytorch☆102Updated 2 years ago
- Implementation of the proposed Adam-atan2 from Google Deepmind in Pytorch☆134Updated 3 months ago
- ResiDual: Transformer with Dual Residual Connections, https://arxiv.org/abs/2304.14802☆97Updated 2 years ago
- Explorations into adversarial losses on top of autoregressive loss for language modeling☆41Updated last month
- Just some miscellaneous utility functions / decorators / modules related to Pytorch and Accelerate to help speed up implementation of new…☆126Updated last year
- Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT☆224Updated last year
- Attempt to make multiple residual streams from Bytedance's Hyper-Connections paper accessible to the public☆163Updated last week
- Implementation of Zorro, Masked Multimodal Transformer, in Pytorch☆98Updated 2 years ago
- Griffin MQA + Hawk Linear RNN Hybrid☆88Updated last year
- Self contained pytorch implementation of a sinkhorn based router, for mixture of experts or otherwise☆40Updated last year
- Implementation of a Light Recurrent Unit in Pytorch☆49Updated last year
- Implementation of the conditionally routed attention in the CoLT5 architecture, in Pytorch☆231Updated last year
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆133Updated 2 months ago
- Triton Implementation of HyperAttention Algorithm☆48Updated 2 years ago
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆103Updated last year
- Implementation of Discrete Key / Value Bottleneck, in Pytorch☆88Updated 2 years ago
- Implementation of an Attention layer where each head can attend to more than just one token, using coordinate descent to pick topk☆47Updated 2 years ago