Skip to content
Longterm Wiki
Index
Grant·6qL5dkV7So·Record·Profile

Grant: Benchmarking and comparing different evaluation awareness metrics (Manifund → Hieu Minh Nguyen)

Verdictpartial50%
1 check · 4/9/2026

Partial deterministic match: amount matched but name did not (20000 rows)

Our claim

entire record
Name
Benchmarking and comparing different evaluation awareness metrics
Amount
$3,000
Currency
USD
Date
August 6, 2025
Notes
[Technical AI safety, AI governance] LLMs often know when they are being evaluated. We’ll do a study comparing various methods to measure and monitor this capability.

Source evidence

1 src · 1 check
partial50%deterministic-row-match · 4/9/2026
Name
We're a team building World-Model-Lens, a library for Interpretability
Slug
were-a-team-building-world-model-lens-a-library-for-interpretability-
Date
2026-03-30T01:31:05.206Z

NotePartial deterministic match: amount matched but name did not (20000 rows)

Case № 6qL5dkV7SoFiled 4/9/2026Confidence 50%
Source Check: Grant: Benchmarking and comparing different evaluation awareness metrics (Manifund -> Hieu Minh Nguyen) | Longterm Wiki