INDEX

Explanations

biological protein and common words

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

This neuron detects punctuation and structural elements that separate different parts in technical contexts, particularly commas, parentheses, and hyphens within code, scientific text, or technical specifications.

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

The neuron fires most strongly on specialized technical nouns or jargon terms (e.g. domain-specific identifiers and class or concept names).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/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

なども

-1.23

 sile

-1.20

？”

-1.16

颪

-1.15

 faldas

-1.13

 också

-1.13

 egentligen

-1.10

躄

-1.06

 افزود

-1.06

 대한

-1.06

POSITIVE LOGITS

他有

1.36

你有

1.22

 realms

1.19

 perbedaan

1.17

for

1.17

 potreb

1.16

 βρί

1.16

 possesses

1.13

 proposte

1.13

though

1.12

Activations Density 0.206%

biological protein and common words

This neuron detects punctuation and structural elements that separate different parts in technical contexts, particularly commas, parentheses, and hyphens within code, scientific text, or technical specifications.

The neuron fires most strongly on specialized technical nouns or jargon terms (e.g. domain-specific identifiers and class or concept names).

No Comments

No Known Activations

biological protein and common words

This neuron detects punctuation and structural elements that separate different parts in technical contexts, particularly commas, parentheses, and hyphens within code, scientific text, or technical specifications.

The neuron fires most strongly on specialized technical nouns or jargon terms (e.g. domain-specific identifiers and class or concept names).

No Comments

No Known Activations