Back to browse
Who's in the weights? – which people 13 language models know

Who's in the weights? – which people 13 language models know

by heterodoxjedi·Jun 22, 2026·1 point·0 comments

AI Analysis

●●●BangerZero to OneBig BrainRabbit Hole

Tests 13 LLMs on 291 people to reveal what's actually baked into model weights.

Strengths
  • Novel methodology comparing confidence scores across 13 different model architectures.
  • Correlates model knowledge against Wikipedia pageviews with interactive scatter plot.
  • Reveals meaningful variation between models on the same person's recognition.
Weaknesses
  • Limited to 291 people — broader dataset would strengthen the findings.
  • No API or programmatic access to run custom queries against the test suite.
Category
Target Audience

ML researchers, AI engineers, people curious about model knowledge

Similar To

Hugging Face Open LLM Leaderboard · LMArena

Similar Projects

AI/ML●●●Banger

Sleeping LLM – A language model that remembers by sleeping

Direct weight editing for persistent memory—MEMIT meets LoRA consolidation with null-space math.

WizardryBig BrainZero to One
vbaranov87
203mo ago