INDEX

Explanations

modelmodel and models

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

頑

-1.67

 acresc

-1.33

 czarny

-1.33

 underpin

-1.32

 Uniwers

-1.32

 niebieski

-1.31

親切

-1.30

tır

-1.26

celana

-1.24

 که

-1.23

POSITIVE LOGITS

这些

1.45

却

1.32

abinieri

1.32

.”

1.26

влено

1.24

="&#

1.24

只是

1.23

 ليس

1.19

GillaGilla

1.17

 هذه

1.16

Activations Density 0.003%