INDEX

Explanations

o w, o, vo sequences

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amina

-0.11

nag

-0.10

 Longer

-0.10

SB

-0.09

 relinqu

-0.09

 expired

-0.09

 Perry

-0.09

 dehydration

-0.09

 Compar

-0.09

-min

-0.09

POSITIVE LOGITS

0.10

 Pavel

0.09

glo

0.09

unf

0.09

uft

0.09

 receptive

0.08

 recept

0.08

 indirect

0.08

itational

0.08

erson

0.08

Activations Density 0.001%