INDEX

Explanations

nd

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mente

-0.31

kaar

-0.28

ulously

-0.28

ities

-0.27

gia

-0.26

nia

-0.26

 mitig

-0.26

azzi

-0.25

communic

-0.25

CONF

-0.24

POSITIVE LOGITS

erton

0.30

crop

0.27

cut

0.26

sett

0.26

éĴº

0.26

pell

0.25

yers

0.25

erval

0.25

sert

0.25

ailed

0.25

Activations Density 0.036%