INDEX

Explanations

make a decision to obtain

religious or directive speech

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

心配

0.23

 stade

0.23

把它

0.23

傘

0.22

쿱

0.22

 zask

0.21

牒

0.21

 definitiv

0.21

 празд

0.21

 végét

0.21

POSITIVE LOGITS

Gaw

0.26

ICAN

0.24

 xvii

0.24

जीवन

0.24

 Oklahoma

0.23

σταν

0.23

Wat

0.23

 आने

0.23

Activations Density 0.008%