Training Transformers with knowledge localization (SGTM)
☆57Jan 11, 2026Updated 9 months ago
Alternatives and similar repositories for selective-gradient-masking
Users that are interested in selective-gradient-masking are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆12Nov 11, 2025Updated 11 months ago
- Investigating the generalization behavior of LM probes trained to predict truth labels: (1) from one annotator to another, and (2) from e…☆34May 23, 2024Updated 2 years ago
- An Adversarial Co-Evolution Framework for LLM Agent Safety☆33Oct 16, 2026Updated 2 weeks ago
- ☆12Sep 11, 2025Updated last year
- Official code for the ICML 2026 paper "Modular Pretraining Enables Access Control" (GRAM).☆44Jul 8, 2026Updated 3 months ago
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- Open Source Replication of Anthropic's Alignment Faking Paper☆61Apr 4, 2025Updated last year
- Awesome tools for interpreting, manipulating the internals of of deep neural networks.☆21Oct 13, 2026Updated 3 weeks ago
- ☆14Apr 30, 2025Updated last year
- ☆16Oct 13, 2025Updated last year
- Training LLMs to Report Their Learned Behaviors☆31Apr 28, 2026Updated 6 months ago
- ☆11Apr 24, 2025Updated last year
- Implementation of paper 'Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference' [NeurIPS'24…☆26Jun 14, 2024Updated 2 years ago
- AEGIS: Automated Error Generation and Attribution for Multi-Agent Systems☆28Aug 17, 2026Updated 2 months ago
- graphpatch is a library for activation patching on PyTorch neural network models.☆22Feb 11, 2025Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- LatentLoop: Scaling Looped MoE Models with Asymmetric Compute Allocation☆16Oct 15, 2026Updated 2 weeks ago
- A curated resource for interactive evaluation: framework, taxonomy, design principles, and benchmark collection.☆24May 22, 2026Updated 5 months ago
- ☆19Jun 3, 2026Updated 5 months ago
- The Assistant Axis is a direction in activation space that captures how "Assistant-like" a model's behavior is. Models can drift away fro…☆182Jan 20, 2026Updated 9 months ago
- Fluid Language Model Benchmarking☆29Sep 16, 2025Updated last year
- Exploring techniques for code refactoring with formal verification☆11Oct 27, 2024Updated 2 years ago
- ☆25Nov 11, 2024Updated last year
- [NeurIPS 2025 Spotlight] Enhancing Training Data Attribution with Representational Optimization☆13Jan 26, 2026Updated 9 months ago
- The official implementation of HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization☆19Mar 7, 2025Updated last year
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- AgenTracer: A Lightweight Failure Attributor for Agentic Systems☆111Aug 27, 2026Updated 2 months ago
- Code for "Unifying Grokking and Double Descent" from the NeurIPS 2022 ML Safety Workshop.☆11Feb 17, 2023Updated 3 years ago
- Sandbox escape benchmark for LLM capability evaluation☆13Apr 29, 2026Updated 6 months ago
- A browser-first, modular LOD chunk streaming engine prototype for progressive world loading.☆12Aug 10, 2026Updated 2 months ago
- TRACE, a framework for turn-aware credit assignment for multi-turn jailbreak optimization☆22Aug 26, 2026Updated 2 months ago
- Source code for the paper "Positional Attention: Expressivity and Learnability of Algorithmic Computation"☆13May 26, 2025Updated last year
- Accompanying codebase for neuroscope.io, a website for displaying max activating dataset examples for language model neurons☆15Feb 13, 2023Updated 3 years ago
- ☆19Updated this week
- Prompts used in the Automated Auditing Blog Post☆172Jul 24, 2025Updated last year
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- Source codes for paper "BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity".☆19Jan 10, 2026Updated 9 months ago
- Official PyTorch implementation for "TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors" [ACL 2026]☆49Apr 14, 2026Updated 6 months ago
- Megatron-LM setup in the smol-cluster☆12Aug 26, 2024Updated 2 years ago
- [ICLR 2025] Official Repository for "Tamper-Resistant Safeguards for Open-Weight LLMs"☆71Jun 9, 2025Updated last year
- Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization☆51Jul 28, 2024Updated 2 years ago
- ☆26May 28, 2025Updated last year
- ☆10Jul 18, 2026Updated 3 months ago