RobertCsordas / moe_layer
sigma-MoE layer
☆18Updated 10 months ago
Related projects ⓘ
Alternatives and complementary repositories for moe_layer
- Official repository for the paper "Approximating Two-Layer Feedforward Networks for Efficient Transformers"☆36Updated 11 months ago
- [NeurIPS 2023] Sparse Modular Activation for Efficient Sequence Modeling☆35Updated 11 months ago
- ☆31Updated 10 months ago
- ☆18Updated 5 months ago
- Reference implementation of "Softmax Attention with Constant Cost per Token" (Heinsen, 2024)☆24Updated 5 months ago
- Fine-Tuning Pre-trained Transformers into Decaying Fast Weights☆19Updated 2 years ago
- ☆44Updated 4 months ago
- My explorations into editing the knowledge and memories of an attention network☆34Updated last year
- ☆21Updated last month
- Here we will test various linear attention designs.☆56Updated 6 months ago
- Official Repository for Efficient Linear-Time Attention Transformers.☆18Updated 5 months ago
- ☆45Updated 9 months ago
- The official repository for our paper "The Dual Form of Neural Networks Revisited: Connecting Test Time Predictions to Training Patterns …☆16Updated 11 months ago
- GoldFinch and other hybrid transformer components☆39Updated 3 months ago
- Blog post☆16Updated 8 months ago
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆24Updated 6 months ago
- Implementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing☆47Updated 2 years ago
- ☆24Updated 8 months ago
- Repository for Skill Set Optimization☆12Updated 3 months ago
- [NeurIPS 2023 spotlight] Official implementation of HGRN in our NeurIPS 2023 paper - Hierarchically Gated Recurrent Neural Network for Se…☆61Updated 6 months ago
- Official code for the paper "Attention as a Hypernetwork"☆23Updated 4 months ago
- 32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.☆42Updated last year
- [ICLR 2023] "Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers" by Tianlong Chen*, Zhenyu Zhang*, Ajay Jaiswal…☆44Updated last year
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆49Updated last year
- RWKV model implementation☆38Updated last year
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆54Updated 3 months ago
- Linear Attention Sequence Parallelism (LASP)☆64Updated 5 months ago
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated 5 months ago
- Official repository for the paper "SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention"☆91Updated last month