INDEX

Explanations

coordinates, dimensions, and formatting

The neuron activates on numeric literals (numbers) in code or markup.

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



-0.91

 that

-0.87

yt

-0.83

aClass

-0.81

Although

-0.80

owo

-0.80

 tratta

-0.80

 appen

-0.79

 terribly

-0.79

gts

-0.79

POSITIVE LOGITS

 новый

1.03

Location

0.98

 новое

0.97

 новом

0.95

anchor

0.93

Point

0.93

point

0.93

POINT

0.93

 данный

0.92

 новую

0.92

Activations Density 0.001%