INDEX

Explanations

identifying similarities and differences

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

任何

-1.41

ruka

-1.32

 kwietnia

-1.23

大约

-1.18

西

-1.17

 Découvrez

-1.17

σκ

-1.16

 antiga

-1.15

 capucha

-1.13

≃

-1.13

POSITIVE LOGITS

 that

2.16

 from

1.59

 their

1.39

if

1.38

 забезпе

1.24

for

1.24

 this

1.16

 presumably

1.15

вного

1.08

 there

1.04

Activations Density 0.535%