ambisinister / mla-experimentsLinks
Experiments on Multi-Head Latent Attention
☆95Updated last year
Alternatives and similar repositories for mla-experiments
Users that are interested in mla-experiments are comparing it to the libraries listed below
Sorting:
- Flash-Muon: An Efficient Implementation of Muon Optimizer☆174Updated 2 months ago
- The evaluation framework for training-free sparse attention in LLMs☆91Updated 2 months ago
- ☆81Updated last year
- ☆123Updated 3 months ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆85Updated last month
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆127Updated last year
- Fast and memory-efficient exact attention☆70Updated 5 months ago
- Transformers components but in Triton☆34Updated 3 months ago
- Code for studying the super weight in LLM☆115Updated 8 months ago
- Unofficial implementation for the paper "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆170Updated last year
- Implementation of the paper: "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆104Updated last week
- An unofficial implementation of "Mixture-of-Depths: Dynamically allocating compute in transformer-based language models"☆35Updated last year
- ☆80Updated 6 months ago
- ☆140Updated 6 months ago
- Low-bit optimizers for PyTorch☆130Updated last year
- Implementation of Infini-Transformer in Pytorch☆111Updated 7 months ago
- ☆149Updated 2 years ago
- ☆118Updated last year
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆144Updated last year
- EE-LLM is a framework for large-scale training and inference of early-exit (EE) large language models (LLMs).☆66Updated last year
- Get down and dirty with FlashAttention2.0 in pytorch, plug in and play no complex CUDA kernels☆107Updated 2 years ago
- The official implementation of the paper "What Matters in Transformers? Not All Attention is Needed".☆174Updated 5 months ago
- An extension of the nanoGPT repository for training small MOE models.☆181Updated 5 months ago
- The official implementation of the paper "Towards Efficient Mixture of Experts: A Holistic Study of Compression Techniques (TMLR)".☆74Updated 5 months ago
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆73Updated last year
- Awesome Triton Resources☆33Updated 4 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆233Updated last week
- ☆55Updated last year
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆165Updated last year
- 🔥 A minimal training framework for scaling FLA models☆233Updated 2 weeks ago