INDEX

Explanations

membership aspects and contexts

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

er

-0.11

Ø±Øµ

-0.10

å·ŀåĮº

-0.10

ness

-0.09

markup

-0.09

 à¹Ĩ

-0.09

atives

-0.09

éĻħ

-0.09

é±¿

-0.09

à¹Ĩ

-0.09

POSITIVE LOGITS

elter

0.11

å¢ħ

0.10

elve

0.09

ilon

0.09

 Ø§ÙĦØ¹Ø±Ø¨ÙĬØ©

0.08

spe

0.08

æĥĩ

0.08

urg

0.08

ç¨·

0.08

ëĭĪëĭ¤

0.08

Activations Density 0.009%