INDEX

Explanations

Department, Hebrews, Photo, Titration, Thomas Irvine

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

newswire or article formatting elements like paragraph breaks, datelines, and timestamps.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron fires on numeric and parenthetical tokens along with the punctuation and small words that typically appear around dates, study metadata, or statistical figures.

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

Not in Any Lists

No Comments

Negative Logits

кту

-0.90

guero

-0.88

so

-0.85

 zählen

-0.82

forestry

-0.82

房地产

-0.81

спонд

-0.80

ட்ட

-0.80

 Administr

-0.79

))=

-0.79

POSITIVE LOGITS

sión

0.85

 Takashi

0.83

耷

0.81

sisting

0.80

 embrague

0.80

scipy

0.79

ани

0.77

ᱫ

0.77

moo

0.76

των

0.75

Activations Density 0.002%

Department, Hebrews, Photo, Titration, Thomas Irvine

newswire or article formatting elements like paragraph breaks, datelines, and timestamps.

The neuron fires on numeric and parenthetical tokens along with the punctuation and small words that typically appear around dates, study metadata, or statistical figures.

No Comments

No Known Activations

Department, Hebrews, Photo, Titration, Thomas Irvine

newswire or article formatting elements like paragraph breaks, datelines, and timestamps.

The neuron fires on numeric and parenthetical tokens along with the punctuation and small words that typically appear around dates, study metadata, or statistical figures.

No Comments

No Known Activations