INDEX

Explanations

what is the questions

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

sibility

-1.53

 jednom

-1.45

 mêmes

-1.41

izhou

-1.40

 incredi

-1.40

 mesmos

-1.37

Joueur

-1.33



-1.32

 kalem

-1.32

ATELY

-1.30

POSITIVE LOGITS

1.71

 reason

1.67

do

1.63

new

1.63

 does

1.56

 difference

1.50

 best

1.47

 about

1.46

 uses

1.45

Activations Density 0.038%