TobiasNorlund / retroLinks
Official repo to On the Generalization Ability of Retrieval-Enhanced Transformers
☆40Updated last year
Alternatives and similar repositories for retro
Users that are interested in retro are comparing it to the libraries listed below
Sorting:
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆77Updated last week
- ☆109Updated last year
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆42Updated last year
- ☆20Updated 2 months ago
- ☆147Updated 2 years ago
- Simple implementation of Speculative Sampling in NumPy for GPT-2.☆95Updated last year
- [NeurIPS'23] Speculative Decoding with Big Little Decoder☆92Updated last year
- ☆51Updated 7 months ago
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆59Updated 8 months ago
- ☆81Updated last year
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆116Updated 6 months ago
- Vocabulary Parallelism☆19Updated 3 months ago
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆210Updated 10 months ago
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆137Updated 10 months ago
- Simple and efficient pytorch-native transformer training and inference (batched)☆76Updated last year
- ☆126Updated last year
- Boosting 4-bit inference kernels with 2:4 Sparsity☆79Updated 9 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆219Updated 6 months ago
- ☆105Updated 10 months ago
- NAACL '24 (Best Demo Paper RunnerUp) / MlSys @ NeurIPS '23 - RedCoast: A Lightweight Tool to Automate Distributed Training and Inference☆66Updated 6 months ago
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆68Updated 10 months ago
- Retrieval as Attention☆82Updated 2 years ago
- ☆38Updated last year
- ☆34Updated 2 years ago
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆70Updated last year
- Layer-Condensed KV cache w/ 10 times larger batch size, fewer params and less computation. Dramatic speed up with better task performance…☆149Updated 2 months ago
- ☆39Updated last year
- ☆130Updated 4 months ago
- ☆74Updated last year
- Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding (EMNLP 2023 Long)☆60Updated 8 months ago