INDEX

Explanations

after sentence terminators

The neuron activates on numeric tokens (especially multi‐digit numbers) in mathematical or technical contexts.

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

řízen

-0.94

resulting

-0.91

#![

-0.91

に使

-0.89

 accueillir

-0.86

teki

-0.86

 parcs

-0.84

∼

-0.84

 villes

-0.83

 quart

-0.83

POSITIVE LOGITS

 with

1.13

for

1.02

 During

0.99

There

0.95

YTA

0.94

 there

0.92

 because

0.92

 While

0.91

 wraz

0.90

 Throughout

0.89

Activations Density 0.064%