5 / papers
Opening the model up — circuits, features, and what "explaining a behaviour" actually means. 5 open-access papers by Pranay Mahendrakar, each with a permanent DOI.
A common reading of the LLM routing literature holds that delegation works without self-knowledge: a router allocates queries using external features of the task and observed outcome statistics over a model pool, and never consults the…
Large language models generate fluent text that is sometimes false, and a substantial literature now proposes to have the model notice and repair those errors while it writes. Parts of the problem are settled. Semantic entropy detects…
State-space models did not merely "open a door" beyond attention; they walked through it. Mamba, Mamba-2, Samba, Jamba, Granite 4, Hymba, and a growing family of hybrid attention-SSM architectures are now in production language models…
In-context learning (ICL) — the ability of a transformer language model to acquire a new input-output mapping from a handful of demonstrations in its prompt, with no weight updates — remains the most striking and least understood…
This paper introduces a novel theoretical framework for understanding semantic processing in Large Language Models through the lens of sheaf theory and quantum contextuality. We propose that the semantic structure of LLM-generated text can…
Pranay Mahendrakar has published 5 open-access papers on Interpretability: "Predicting Your Own Failure Is Not Predicting Another Model's Advantage", "The Self-Verification Gap", "Memory Architectures Beyond Attention", "Mechanistic Interpretability of In-Context Learning" and "Sheaf-Theoretic Semantics and Quantum Contextuality in Large Language Models". Each is deposited on Zenodo with a permanent DOI and is free to read.
