INDEX

Explanations

comma

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĲĦæł·

-0.27

 particular

-0.27

çī¹å®ļ

-0.27

imiento

-0.25

ids

-0.25

urrent

-0.24

äº

-0.24

å±ĭåŃĲ

-0.24

iders

-0.24

ury

-0.24

POSITIVE LOGITS

arella

0.26

sell

0.25

çķħéĶĢ

0.24

=params

0.24

 exon

0.24

 Phillips

0.24

gene

0.23

à¹Ģà¸ŀà¸¥à¸ĩ

0.23

ç¢İ

0.23

ise

0.23

Activations Density 0.019%