INDEX

Explanations

paid use chance record expressed

The neuron flags descriptive adjectives or past‐participle forms (and similar adjectival tokens) that characterize states or properties.

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

to

-1.79

its

-1.21

-1.16

 enceinte

-1.15

But

-1.14

but

-1.12

of

-1.09

 located

-1.09

 vall

-1.08

hallo

-1.05

POSITIVE LOGITS

 esimerkiksi

1.45

 kysy

1.42

 ensimmä

1.34

 siinä

1.30

 شنبه

1.29

jścia

1.29

 erity

1.26

 pysty

1.26

 muuta

1.25

kać

1.25

Activations Density 0.128%