INDEX

Explanations

before

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

erna

-0.09

reme

-0.08

 Boyd

-0.08

amas

-0.08

TEM

-0.08

aki

-0.08

alar

-0.08

uni

-0.08

eren

-0.08

ks

-0.08

POSITIVE LOGITS

 previously

0.49

 previous

0.39

 Previously

0.37

Previously

0.35

 Previous

0.29

previous

0.28

Previous

0.28

 before

0.27

 earlier

0.23

 past

0.22

Activations Density 0.375%