INDEX

Explanations

No Explanations Found

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

çĽ´è¾¾

-0.31

è®°å½ķ

-0.27

ection

-0.26

åĽºå®ļ

-0.26

å®ŀ

-0.25

éĩĩ

-0.25

éĩįå¤į

-0.25

çĽ´æİ¥

-0.25

 Natal

-0.25

 Direct

-0.24

POSITIVE LOGITS

anes

0.25

 hammered

0.24

-method

0.24

 stripper

0.24

äºĨä¸įå°ĳ

0.24

èĪªçº¿

0.24

 slew

0.24

ä¼º

0.23

MethodName

0.23

.localization

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.