INDEX

Explanations

combinations of

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

éļ¼

-0.24

ä¸Ģæ¨£

-0.23

ctic

-0.23

ERICA

-0.23

TypeName

-0.22

 similarly

-0.22

åĲĮæł·æĺ¯

-0.21

å·¥åķĨå±Ģ

-0.21

ARAM

-0.21

ä¸Ģæł·çļĦ

-0.21

POSITIVE LOGITS

 Ð¿ÑĢÐ¸ÑıÑĤÐ½

0.26

ittest

0.23

 Recorder

0.23

å°ģ

0.22

 spokesperson

0.22

 unmist

0.22

ðŁĵĥ

0.22

amp

0.21

Õµ

0.21

itÃł

0.21

Activations Density 0.429%