41,554 blogs · 5,225,211 posts Claim your blog

Record · corpus.blog/posts/01a0b653-445d-7299-8b28-ad1ce5375914

In Progress -- Applications of Wikidata For LLM Evaluation and Automatic Mechinterp Validation: A Pilot Study and Commentary

dactile.net · published 17 November 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at dactile.net Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
External links: 9
Words captured: 1,691

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

Toy Models of Superposition Read at the source ↗

transformer-circuits.pub

Towards Monosemanticity: Decomposing Language Models With Dictionary Learning Read at the source ↗

transformer-circuits.pub

https://arxiv.org/abs/2309.08600

arxiv.org

Sparse Crosscoders for Cross-Layer Features and Model Diffing Read at the source ↗

transformer-circuits.pub

https://openaipublic.blob.core.windows.net/neuron-explainer/paper/index.html

openaipublic.blob.core.windows.net

https://arxiv.org/abs/2405.07436

arxiv.org

https://www.wikidata.org/wiki/Wikidata:Data_model

wikidata.org

https://alexwlchan.net/til/2023/what-is-a-snak/

alexwlchan.net

https://www.wikidata.org/wiki/Q276525

wikidata.org

Elsewhere on dactile.net (5)

Dev Notes 08: Initial Explorations with wikidata Read at the source ↗

https://dactile.net/p/dev-notes-09

dactile.net

Dev Notes 10: Progress In Wikidata Processing to Study LLMs Ability To Articulate Connections Read at the source ↗

Dev Notes 11: Progress In Identifying Connected Entity Sets for LLM Evaluation Read at the source ↗

Dev Notes 12: Blog Improvements and Playing With Wikidata Prompts Read at the source ↗