Efficient 2:4 sparse training algorithms and implementations
☆63Dec 8, 2024Updated last year
Alternatives and similar repositories for 2by4-pretrain
Users that are interested in 2by4-pretrain are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Code for "Accelerating Transformer Pre-training with 2:4 Sparsity"☆28Dec 8, 2024Updated last year
- [ICML 2026] Residual Context Diffusion (RCD): Repurposing discarded signals as structured priors for high-performance reasoning in dLLMs.☆62Jun 28, 2026Updated 4 months ago
- Official implementation for "Pruning Large Language Models with Semi-Structural Adaptive Sparse Training" (AAAI 2025)☆20Jul 1, 2025Updated last year
- A selective knowledge distillation algorithm for efficient speculative decoders☆40Nov 27, 2025Updated 11 months ago
- ☆64Jul 21, 2024Updated 2 years ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Fast dLLM pretraining attention implementation☆15Oct 14, 2026Updated 3 weeks ago
- Implementation of NM sparsity recipe presented in the paper "Progressive Gradient Flow for Robust N:M Sparsity Training in Transformers".☆11Feb 5, 2024Updated 2 years ago
- ☆245Nov 9, 2022Updated 3 years ago
- Combining SOAP and MUON☆24Feb 11, 2025Updated last year
- Code Repository for the NeurIPS 2024 Paper "Toward Efficient Inference for Mixture of Experts".☆19Oct 30, 2024Updated 2 years ago
- ☆173Feb 15, 2025Updated last year
- [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.☆123Dec 20, 2024Updated last year
- An experiment to see if chatgpt can improve the output of the stanford alpaca dataset☆12Mar 29, 2023Updated 3 years ago
- ☆11Dec 26, 2025Updated 10 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- The loss landscape of Large Language Models resemble basin!☆49Jul 8, 2025Updated last year
- Implementation of experiments from The No Free Lunch Theorem, Kolmogorov Complexity, and the Role of Inductive Biases in Machine Learning☆17May 14, 2023Updated 3 years ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆269Aug 9, 2025Updated last year
- ☆158Jun 22, 2023Updated 3 years ago
- A family of efficient edge language models in 100M~1B sizes.☆20Feb 14, 2025Updated last year
- Work in progress.☆82Nov 25, 2025Updated 11 months ago
- ☆14Jul 24, 2024Updated 2 years ago
- Source codes for paper "BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity".☆19Jan 10, 2026Updated 9 months ago
- [ICLR 2026] SERE: Similarity-Based Expert Re-routing for Efficient Batch Decoding in MoE Models☆23Feb 4, 2026Updated 9 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- [NeurIPS 24 Spotlight] MaskLLM: Learnable Semi-structured Sparsity for Large Language Models☆192Jan 1, 2025Updated last year
- Code implementation of DuoGPT (NeurIPS 2025, https://arxiv.org/abs/2506.20194)☆10Sep 23, 2025Updated last year
- ☆16Aug 14, 2026Updated 2 months ago
- A lightweight execution guard for running LLM-generated Python in a normal Python process☆12Oct 7, 2026Updated last month
- A simple gitlab/github web hooks daemon☆16Oct 22, 2026Updated 2 weeks ago
- [CVPR'24] Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression☆17Jul 1, 2024Updated 2 years ago
- ☆28Jul 2, 2026Updated 4 months ago
- [ICLR 2026] The official implementation of "PTQ4ARVG: Post-Training Quantization for AutoRegressive Visual Generation Models"☆18Feb 3, 2026Updated 9 months ago
- 清华大学研究生社会实践系统爬虫☆18Jun 4, 2024Updated 2 years ago
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- ORL: On-Policy Residual Learning for Diffusion Language Models☆27Oct 16, 2026Updated 3 weeks ago
- 2020年秋国科大图像处理与分析（彭思龙、杨戈）编程课后作业☆10Feb 3, 2021Updated 5 years ago
- code for the paper "A Statistical Framework for Low-bitwidth Training of Deep Neural Networks"☆29Oct 31, 2020Updated 6 years ago
- [NeurIPS 2024] Official implementation of "Grid4D: 4D Decomposed Hash Encoding for High-Fidelity Dynamic Gaussian Splatting"☆95Jul 27, 2025Updated last year
- [OSDI 2025] DecDEC: A Systems Approach to Advancing Low‑Bit LLM Quantization☆27Oct 13, 2026Updated 3 weeks ago
- APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decoding☆14Jul 22, 2024Updated 2 years ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆130Jul 13, 2024Updated 2 years ago