conftrace_

mechanistic interpretability

133 papers

Explore in graph

Also known as

MI

Co-occurring keywords

large language model (13587) language model (4599) sparse autoencoder (112) attention head (146) neural network (6616) circuit analysis (17) in-context learning (1405) circuit discovery (14) attention mechanism (4001) activation patching (23)

Papers

What Makes and Breaks Safety Fine-tuning? A Mechanistic Study NIPS 2024
Finding Transformer Circuits With Edge Pruning NIPS 2024
Unlocking the Future: Exploring Look-Ahead Planning Mechanistic Interpretability in Large Language Models EMNLP 2024
How does GPT-2 Predict Acronyms? Extracting and Understanding a Circuit via Mechanistic Interpretability AISTATS 2024
Talking Heads: Understanding Inter-Layer Communication in Transformer Language Models NIPS 2024
Hypothesis Testing the Circuit Hypothesis in LLMs NIPS 2024
Transformers Represent Belief State Geometry in their Residual Stream NIPS 2024
InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques NIPS 2024
Towards Interpretable Sequence Continuation: Analyzing Shared Circuits in Large Language Models EMNLP 2024
Mechanistic? EMNLP 2024
Copy Suppression: Comprehensively Understanding a Motif in Language Model Attention Heads EMNLP 2024
Deeper Insights Without Updates: The Power of In-Context Learning Over Fine-Tuning EMNLP 2024
Where does In-context Learning Happen in Large Language Models? NIPS 2024
LLM Circuit Analyses Are Consistent Across Training and Scale NIPS 2024
Compact Proofs of Model Performance via Mechanistic Interpretability NIPS 2024
Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning NIPS 2024
Beyond the Doors of Perception: Vision Transformers Represent Relations Between Objects NIPS 2024
Preference Tuning For Toxicity Mitigation Generalizes Across Languages EMNLP 2024
On the Similarity of Circuits across Languages: a Case Study on the Subject-verb Agreement Task EMNLP 2024
Information Flow Routes: Automatically Interpreting Language Models at Scale EMNLP 2024
The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis EMNLP 2024
Monotonic Representation of Numeric Attributes in Language Models ACL 2024
Detecting and Understanding Vulnerabilities in Language Models via Mechanistic Interpretability IJCAI 2024
Scale Alone Does not Improve Mechanistic Interpretability in Vision Models NIPS 2023
The Clock and the Pizza: Two Stories in Mechanistic Explanation of Neural Networks NIPS 2023