INDEX

Explanations

numbers, symbols, or foreign words

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

This neuron fires strongly on words that begin new discourse segments (e.g. the first word of a sentence or paragraph).

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

periods following sentences, especially at major sentence or paragraph boundaries.

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

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

大多數

0.69

大多数

0.68

 உணர

0.67

大部分

0.60

 대부분

0.60

 изначально

0.59

 considérer

0.57

最初に

0.57

 plaus

0.57

主張

0.57

POSITIVE LOGITS

 fútbol

0.82

 انتخابات

0.76

 फुटबॉल

0.75

 anunció

0.74

 música

0.73

 नए

0.72

 soccer

0.72

 автомоби

0.72

 футбо

0.71

 فوتبال

0.71

Activations Density 20.470%

numbers, symbols, or foreign words

This neuron fires strongly on words that begin new discourse segments (e.g. the first word of a sentence or paragraph).

periods following sentences, especially at major sentence or paragraph boundaries.

No Comments

No Known Activations

numbers, symbols, or foreign words

This neuron fires strongly on words that begin new discourse segments (e.g. the first word of a sentence or paragraph).

periods following sentences, especially at major sentence or paragraph boundaries.

No Comments

No Known Activations