INDEX

Explanations

contrast or qualification

not X

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

0.35

oth

0.33

 OTHER

0.32

 ihrer

0.32

 ఇతర

0.32

 অন্যান্য

0.31

us

0.31

 इतर

0.30

{\

0.30

">⁢</

0.30

POSITIVE LOGITS

 heartache

0.38

 heartbreak

0.35

 paradox

0.35

 farce

0.35

 dogma

0.34

机遇

0.34

 عین

0.33

Hyper

0.33

кономі

0.32

缛

0.32

Activations Density 0.338%