Stars
Training Sparse Autoencoders on Language Models
A framework for few-shot evaluation of language models.
Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
wolfecameron / nanoMoE
Forked from karpathy/nanoGPTAn extension of the nanoGPT repository for training small MOE models.
Fast and memory-efficient exact attention
High quality training free inpaint for every stable diffusion model. Supports ComfyUI
Small and quick script, that puts finished magisk files to waydroid directory. Fully working Magisk Delta!!
⚡ TabPFN: Foundation Model for Tabular Data ⚡
Can Knowledge Editing Really Correct Hallucinations? (ICLR 2025)
A Unified Framework for High-Performance and Extensible LLM Steering
AnyEdit: Edit Any Knowledge Encoded in Language Models, ICML 2025
AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models, ICLR 2025 (Outstanding Paper)
KOWX712 / PlayIntegrityFix
Forked from jyotidwi/PlayIntegrityFixFix Play Integrity verdicts.
The official implementation for [NeurIPS2025 Oral] Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free
Unofficial implementation of Titans, SOTA memory for transformers, in Pytorch
Official implementation of paper "Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models"
Steering Llama 2 with Contrastive Activation Addition
Integrating neurosymbolic representations into LLMs for interpretability, steering, and running symbolic algorithms
🚀 One-stop solution for creating your digital avatar from chat history 💡 Fine-tune LLMs with your chat logs to capture your unique style, then bind to a chatbot to bring your digital self to life. …
Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.
Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.
Vector (and Scalar) Quantization, in Pytorch
MemGen: Weaving Generative Latent Memory for Self-Evolving Agents