INDEX

Explanations

article + diverse noun/adjective

This neuron flags high‐frequency “function” tokens—especially determiners and common participial/linking words.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-2.20

 عندما

-1.91

 неболь

-1.91

-1.88

𝐭

-1.85

if

-1.83

-1.80

 این

-1.77

ꈊ

-1.73

～！

-1.70

POSITIVE LOGITS

damento

2.09

 بريط

2.05

出てくる

1.93

 休闲

1.92

 nahil

1.91

seded

1.89



1.77

なった

1.77

mpagne

1.74

 别墅

1.73

Activations Density 0.002%