INDEX

Explanations

entity names and subsequent context

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

are

-1.65

 kolei

-1.31

üyü

-1.23

Как

-1.20

 offer

-1.19

Kı

-1.18

şu

-1.18

 दौरान

-1.18

 potrze

-1.18

 dziś

-1.18

POSITIVE LOGITS

 after

1.95

 when

1.86

for

1.77

 until

1.58

to

1.48

if

1.47

by

1.43

 they

1.41

 instead

1.41

«

1.39

Activations Density 0.247%