INDEX

Explanations

The neuron selectively activates on numeric tokens (digits and numbers, often with decimals or negative signs).

numbers and symbols

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

 äls

-1.02

 puts

-0.97

utf

-0.95

gr

-0.93

no

-0.93

ro

-0.92

冈

-0.92

-0.91

min

-0.91

 will

-0.91

POSITIVE LOGITS

 raiſ

1.43

 Reſ

1.38

ſten

1.27



1.23

 Livro

1.18

 Diſ

1.16

ﾐ

1.15

 Melhor

1.13

 leſs

1.13

 tatuagem

1.12

Activations Density 0.196%