INDEX

Explanations

foreign languages and programming

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

words in non-English languages, particularly Romance languages and Asian languages.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

This neuron fires on tokens written in non-basic-ASCII scripts or with non-English characters (e.g. Chinese characters, accented letters, or other non-Latin symbols).

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

 were

-3.45

-2.98

 mereka

-2.77

 predominantly

-2.58

 principalmente

-2.56

-2.55

-2.50

 dando

-2.41

-2.36

POSITIVE LOGITS

峮

2.86

ﻒ

2.78

fiées

2.70

 priser

2.67

᭝

2.55

した後

2.48

 зелё

2.48

埸

2.47

ꗃ

2.45

㈬

2.44

Activations Density 0.001%

foreign languages and programming

words in non-English languages, particularly Romance languages and Asian languages.

This neuron fires on tokens written in non-basic-ASCII scripts or with non-English characters (e.g. Chinese characters, accented letters, or other non-Latin symbols).

No Comments

No Known Activations

foreign languages and programming

words in non-English languages, particularly Romance languages and Asian languages.

This neuron fires on tokens written in non-basic-ASCII scripts or with non-English characters (e.g. Chinese characters, accented letters, or other non-Latin symbols).

No Comments

No Known Activations