7 papers
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones
Daking Rai, Samuel Miller, Kevin Moran +1
Despite remarkable advances in coding capabilities, language models (LMs) still struggle with simple syntactic tasks such as generating balanced parentheses. In this study, we inve…
Data-driven Circuit Discovery for Interpretability of Language Models
Daking Rai, Mor Geva, Ziyu Yao
Circuit discovery aims to explain how language models (LMs) implement a specific task by localizing and interpreting a circuit, a computational subgraph responsible for the LM's be…
A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
Daking Rai, Yilun Zhou, Shi Feng +2
Mechanistic interpretability (MI) is an emerging sub-field of interpretability that seeks to understand a neural network model by reverse-engineering its internal computations. Rec…
A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models
Dong Shu, Xuansheng Wu, Haiyan Zhao +4
Large Language Models (LLMs) have transformed natural language processing, yet their internal mechanisms remain largely opaque. Recently, mechanistic interpretability has attracted…
All for One: LLMs Solve Mental Math at the Last Token With Information Transferred From Other Tokens
Siddarth Mamidanna, Daking Rai, Ziyu Yao +1
Large language models (LLMs) demonstrate proficiency across numerous computational tasks, yet their inner workings remain unclear. In theory, the combination of causal self-attenti…
Mechanistic Understanding of Language Models in Syntactic Code Completion
Samuel Miller, Daking Rai, Ziyu Yao
Recently, language models (LMs) have shown impressive proficiency in code generation tasks, especially when fine-tuned on code-specific datasets, commonly known as Code LMs. Howeve…