Pith. sign in

REVIEW 2 cited by

Evaluating Neuron Interpretation Methods of NLP Models

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 2301.12608 v2 pith:KW563D2C submitted 2023-01-30 cs.CL

classification cs.CL
keywords methodsevaluationneuronframeworkinterpretationmethodworkacross
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Neuron Interpretation has gained traction in the field of interpretability, and have provided fine-grained insights into what a model learns and how language knowledge is distributed amongst its different components. However, the lack of evaluation benchmark and metrics have led to siloed progress within these various methods, with very little work comparing them and highlighting their strengths and weaknesses. The reason for this discrepancy is the difficulty of creating ground truth datasets, for example, many neurons within a given model may learn the same phenomena, and hence there may not be one correct answer. Moreover, a learned phenomenon may spread across several neurons that work together -- surfacing these to create a gold standard challenging. In this work, we propose an evaluation framework that measures the compatibility of a neuron analysis method with other methods. We hypothesize that the more compatible a method is with the majority of the methods, the more confident one can be about its performance. We systematically evaluate our proposed framework and present a comparative analysis of a large set of neuron interpretation methods. We make the evaluation framework available to the community. It enables the evaluation of any new method using 20 concepts and across three pre-trained models.The code is released at https://github.com/fdalvi/neuron-comparative-analysis

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. The Process of Categorical Clipping at the Core of the Genesis of Concepts in Synthetic Neural Cognition

    cs.AI 2025-01 conditional novelty 4.0 of 10

    Words that strongly activate both a lower-layer neuron and its strongly connected upper-layer neuron in GPT-2XL form more semantically similar clusters, which the paper interprets as a clipping process.

  2. How Do Artificial Intelligences Think? The Three Mathematico-Cognitive Factors of Categorical Segmentation Operated by Synthetic Neurons

    q-bio.NC 2024-12 reject novelty 2.0 of 10

    The paper names three components of a neuron's aggregation function as cognitive factors and reports near-unity correlations in GPT-2XL, but the effects are largely true by construction.

Pith tools