ducdauge / sft-llmLinks
Scaling Sparse Fine-Tuning to Large Language Models
☆17Updated last year
Alternatives and similar repositories for sft-llm
Users that are interested in sft-llm are comparing it to the libraries listed below
Sorting:
- Awesome Triton Resources☆33Updated 3 months ago
- [ICLR 2023] "Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers" by Tianlong Chen*, Zhenyu Zhang*, Ajay Jaiswal…☆54Updated 2 years ago
- Code for paper "Patch-Level Training for Large Language Models"☆86Updated 9 months ago
- ☆33Updated last year
- [NAACL 2025] A Closer Look into Mixture-of-Experts in Large Language Models☆54Updated 6 months ago
- The official implementation for Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free☆46Updated 3 months ago
- The open-source materials for paper "Sparsing Law: Towards Large Language Models with Greater Activation Sparsity".☆25Updated 9 months ago
- [NeurIPS 2023] Sparse Modular Activation for Efficient Sequence Modeling☆39Updated last year
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- [EMNLP 2023]Context Compression for Auto-regressive Transformers with Sentinel Tokens☆25Updated last year
- Repository of the paper "Accelerating Transformer Inference for Translation via Parallel Decoding"☆119Updated last year
- Code for "Everybody Prune Now: Structured Pruning of LLMs with only Forward Passes"☆28Updated last year
- ☆20Updated last year
- ☆19Updated 2 years ago
- PyTorch implementation of StableMask (ICML'24)☆14Updated last year
- ☆56Updated last year
- Official implementation of the ICML 2024 paper RoSA (Robust Adaptation)☆42Updated last year
- [ICML 2024] When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Models☆33Updated last year
- Reference implementation of "Softmax Attention with Constant Cost per Token" (Heinsen, 2024)☆24Updated last year
- The source code of "Merging Experts into One: Improving Computational Efficiency of Mixture of Experts (EMNLP 2023)":☆38Updated last year
- Stick-breaking attention☆59Updated last month
- 32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.☆49Updated 2 years ago
- Here we will test various linear attention designs.☆62Updated last year
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆84Updated last month
- ☆85Updated last year
- Using FlexAttention to compute attention with different masking patterns☆44Updated 11 months ago
- The official repository for SkyLadder: Better and Faster Pretraining via Context Window Scheduling☆33Updated 3 weeks ago
- User-friendly implementation of the Mixture-of-Sparse-Attention (MoSA). MoSA selects distinct tokens for each head with expert choice rou…☆25Updated 3 months ago
- ☆51Updated 2 months ago
- Code for "RSQ: Learning from Important Tokens Leads to Better Quantized LLMs"☆18Updated 2 months ago