INDEX

Explanations

responsibility to ensure

contrastive conjunctions and punctuation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

li

0.52

0.44

 Panasonic

0.43

kal

0.43

ani

0.42

ai

0.42

 Smart

0.42

 kalam

0.42

POSITIVE LOGITS

觎

0.52

otard

0.51

레이

0.49

stargo

0.48

ಸರ್

0.48

পর

0.48

 religieux

0.47

рьох

0.47

 ತೊ

0.47

NDAY

0.47

Activations Density 0.001%