Activation Flow: Manufacturing Activations for Steering
cs.LG, cs.AI, cs.CL, math.DG
Submitted: 2026-09-26
Updated: 2026-10-02
Code: https://github.com/HK-Tan/actflow
Terminology
Sources
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- The Llama 3 Herd of Models
- Mistral 7B
- Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models
- Reference-Grafting Matches Fine-Tuning at Eliciting Sandbagged Capabilities
- Mechanistically Eliciting Latent Behaviors in Language Models
- Steered LLM Activations are Non-Surjective
- Qwen2.5 Technical Report
- The Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology
- A Causal Model for Locating and Unlocking Sandbagging in Model Organisms
- Steering Language Models With Activation Engineering
- On the Non-Identifiability of Steering Vectors in Large Language Models
- FishBack: Pullback Fisher Geometry for Optimal Activation Steering in Transformers
- Relational Composition in Neural Networks: A Survey and Call to Action
- Representation Engineering: A Top-Down Approach to AI Transparency
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks