Pith. sign in

REVIEW 2 cited by

Backward Lens: Projecting Language Model Gradients into the Vocabulary Space

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.12865 v1 pith:NJADG56M submitted 2024-02-20 cs.CL cs.AIcs.LG

classification cs.CLcs.AIcs.LG
keywords backwardgradientsinformationforwardlanguagemethodsmodelspass
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Understanding how Transformer-based Language Models (LMs) learn and recall information is a key goal of the deep learning community. Recent interpretability methods project weights and hidden states obtained from the forward pass to the models' vocabularies, helping to uncover how information flows within LMs. In this work, we extend this methodology to LMs' backward pass and gradients. We first prove that a gradient matrix can be cast as a low-rank linear combination of its forward and backward passes' inputs. We then develop methods to project these gradients into vocabulary items and explore the mechanics of how new information is stored in the LMs' neurons.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Verbalizable Representations Form a Global Workspace in Language Models

    cs.CL 2026-07 conditional novelty 7.0 of 10

    Language models represent their current reasoning in a small, readable set of verbalizable vectors (the J-space) that functions like a global workspace.

  2. Dissecting Bias in LLMs: A Mechanistic Interpretability Perspective

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Bias in GPT-2 and Llama-2 is localized to a small set of edges, and ablation of those edges reduces bias while impairing unrelated NLP tasks.

Pith tools