INDEX

Explanations

medical and technical terms

np_acts-logits-general · gemini-2.5-flash-lite

scientific and technical terminology with high specificity.

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron strongly activates on low‐frequency, content‐rich terms (especially technical, domain‐specific, or otherwise uncommon words).

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

∅

-1.05

dig

-1.00

酲

-0.96

)$.

-0.94

Force

-0.94

∽

-0.93

 цвете

-0.92

￡

-0.89

幌

-0.88



-0.88

POSITIVE LOGITS

 seçenek

1.05

 проєкту

0.94

 mitään

0.91

 ambayo

0.90

褚

0.89

 entlang

0.89

際には

0.89

 antwoord

0.87

⇡

0.86

ⲣ

0.86

Activations Density 0.068%

medical and technical terms

scientific and technical terminology with high specificity.

The neuron strongly activates on low‐frequency, content‐rich terms (especially technical, domain‐specific, or otherwise uncommon words).

No Known Activations

medical and technical terms

scientific and technical terminology with high specificity.

The neuron strongly activates on low‐frequency, content‐rich terms (especially technical, domain‐specific, or otherwise uncommon words).

No Known Activations