INDEX

Explanations

referencing above code or link

This neuron activates on Spanish-language words or phrases in the text.

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

of

-1.34

nue

-1.31

 itin

-1.26

 gustar

-1.23

iliger

-1.20

はこんな感じ

-1.18

 univers

-1.17

女

-1.15

 wedi

-1.13

zit

-1.13

POSITIVE LOGITS

opheles

1.43

 étanche

1.38

 dunklen

1.24

 жов

1.22

 fifth

1.22

 pulito

1.18

 відді

1.16

 sportivo

1.13

 know

1.10

 erila

1.10

Activations Density 0.062%