INDEX

Explanations

foreign words

The neuron predominantly activates on numeric tokens or measurements (e.g., “6,” “81.5,” “124.5,” etc.).

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

-1.33

？

-1.22

百合

-1.16

 grama

-1.15

 ❤️❤️

-1.14

勇者

-1.13

 tragé

-1.12

魔女

-1.11

 ★★★

-1.11

❉

-1.10

POSITIVE LOGITS

Variante

1.51

久し

1.44

 mojej

1.43

昔から

1.41

 görüntüsü

1.40

 goede

1.38

 duvar

1.37

쐈

1.35

 těchto

1.35

 cuarentena

1.35

Activations Density 0.027%