INDEX

Explanations

questions and answers

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

 impul

-1.13

䛗

-1.13

 نیز

-1.12

 neces

-1.12

ſſen

-1.11

ttää

-1.09

zembro

-1.09

dzić

-1.07

 देख

-1.04

ofType

-1.04

POSITIVE LOGITS

 answer

1.26

Answer

1.18

 abitanti

1.16

 asked

1.16

 chiesto

1.15

 Answer

1.14

 beträ

1.13

大丈夫です

1.12

 Gefäng

1.12

 kämp

1.08

Activations Density 0.028%