INDEX

Explanations

be + adjective/past participle

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

omet

-0.10

 Soda

-0.10

be

-0.10

nek

-0.10

ocr

-0.10

Aw

-0.10

elf

-0.09

nick

-0.09

not

-0.09

-names

-0.09

POSITIVE LOGITS

aker

0.10

stride

0.09

elor

0.09

anking

0.09

 wise

0.08

alamat

0.08

avers

0.08

&T

0.08

slash

0.08

 kinky

0.08

Activations Density 0.115%