INDEX

Explanations

pronouns followed by actions

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

 that

-2.53

 during

-2.23

 while

-2.09

 exquisitely

-2.09

 despite

-2.08

 some

-2.02

 throughout

-2.00

 superbly

-2.00

 because

-1.98

 although

-1.98

POSITIVE LOGITS

 delightfully

2.30

2.28

 wondrous

2.25

 lauded

2.22

 quirky

2.17

۳

2.16

 dismal

2.16

 impreso

2.13

 decidedly

2.09



2.09

Activations Density 0.090%