INDEX

Explanations

below

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

à¤Ĭ

-0.30

çļĦæĸĩç«ł

-0.26

è²¼

-0.26

çĽĺçĤ¹

-0.25

IID

-0.25

åĲ¬äºĨ

-0.25

-reviewed

-0.25

urls

-0.25

 listed

-0.24

æī£

-0.23

POSITIVE LOGITS

 Sinai

0.27

afen

0.26

uable

0.26

è¿ģ

0.26

extField

0.25

ATUS

0.25

AKER

0.25

fÃ¤llig

0.25

 Condition

0.24

onn

0.24

Activations Density 0.010%