INDEX

Explanations

as ... as comparisons of degree

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

of

-1.98

for

-1.90

ighed

-1.66

by

-1.63

ier

-1.54

Tämä

-1.53

vier

-1.52

 partai

-1.52

 into

-1.50

cels

-1.50

POSITIVE LOGITS

iência

1.51

ሎ

1.45

 folosit

1.45

 nostru

1.44

زارش

1.44

 Collège

1.43



1.41

وانه

1.36

lapping

1.33

 Lycée

1.33

Activations Density 0.029%