INDEX

Explanations

Object.prototype.toString

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

the phrase "suffice it to say" and related formal transitional expressions.

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

This neuron selectively fires on natural‐language (prose) words—especially in a phrase like “suffice it to say” or ordinary English adverbs and nouns—while ignoring code tokens.

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

was

-2.41

is

-1.92

 showed

-1.91

 there

-1.89

Some

-1.88

 should

-1.84

When

-1.84

 used

-1.80

not

-1.78

 because

-1.72

POSITIVE LOGITS

 الآخر

1.76

to

1.75

 indiquer

1.70

めない

1.70

 كافة

1.66

⅘

1.59

fillment

1.55

ми

1.54

 zove

1.54

 něj

1.53

Activations Density 0.044%

Object.prototype.toString

the phrase "suffice it to say" and related formal transitional expressions.

This neuron selectively fires on natural‐language (prose) words—especially in a phrase like “suffice it to say” or ordinary English adverbs and nouns—while ignoring code tokens.

No Comments

No Known Activations

Object.prototype.toString

the phrase "suffice it to say" and related formal transitional expressions.

This neuron selectively fires on natural‐language (prose) words—especially in a phrase like “suffice it to say” or ordinary English adverbs and nouns—while ignoring code tokens.

No Comments

No Known Activations