INDEX

Explanations

index, offset, position

The neuron primarily detects numeric literals (numbers) in code.

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

-1.27

droje

-1.21

nitzer

-1.19

ables

-1.16

-1.13

 Newtonsoft

-1.10

торая

-1.09

xd

-1.08

と感じ

-1.08

してい

-1.07

POSITIVE LOGITS

隨著

1.35

 休闲

1.27

随着

1.23

今年も

1.22

女優

1.17

俳優

1.16

آیا

1.16

因為

1.15

“

1.13

ขอ

1.13

Activations Density 0.033%