INDEX

Explanations

sports brands

The main thing this neuron does is detect sportswear and athletic-gear brand names (often marked with trademarks).

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

."

-1.56

半个

-1.31

-1.23

):

-1.22

-1.19

殯

-1.18

网红

-1.17

 that

-1.16

榈

-1.16

 texte

-1.14

POSITIVE LOGITS

༣

1.37

哙

1.37

 aantal

1.32

 mahd

1.32

 フリル

1.27

 Lainnya

1.27

łgorzata

1.25

 vuelven

1.23

 erhi

1.22

ργασ

1.21

Activations Density 0.016%