Beyond Static Interpretability: Anticipating Post-SFT Mechanisms from Pre-SFT Parameters for Better Tuning
cs.LG, cs.AI, cs.CL
Submitted: 2026-08-25
Updated: 2026-08-25
Comments: 25 pages
Code: https://github.com/Zodiark-ch/Future_localization
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- How to use and interpret activation patching
- Mistral 7B
- Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
- A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
- Open Problems in Mechanistic Interpretability
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Qwen3 Technical Report
- Local Conflict, Targeted Repair: A Perturbation Analysis of Multi-Domain RL
- A Comprehensive Study of Knowledge Editing for Large Language Models
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks