INDEX

Explanations

academic qualifications and professional titles

concepts related to ethics, consent, and attention

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

大当たり

0.46

 disastrous

0.45

弐

0.45

 overcrowded

0.42

すぎて

0.40

elajaran

0.40

 conspired

0.39

安い

0.39

তান্ত্রিক

0.39

прос

0.39

POSITIVE LOGITS

crit

0.46

 बढ़ना

0.44

 bør

0.43

published

0.42

நர்

0.42

 völl

0.42

 doar

0.41

 cuánt

0.40

stüt

0.40

રિક

0.40

Activations Density 0.003%