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

uzzi

-0.34

æī¿åĬŀ

-0.31

]+"

-0.28

åħ³éĹ¨

-0.27

iry

-0.26

ä½Ĩåį´

-0.26

itures

-0.25

ä¿±ä¹Ĳ

-0.25

å¤§éĺŁ

-0.25

RowAt

-0.25

POSITIVE LOGITS

ana

0.25

oyo

0.25

oko

0.25

aka

0.25

`,

0.25

san

0.25

 Sites

0.25

 Helpful

0.24

 Ð½Ð¾Ð²ÑĭÐµ

0.24

hon

0.24

Activations Density 0.015%