INDEX

Explanations

States

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

ANTED

-0.25

åľŁèĢ³

-0.24

opportunitÃł

-0.23

putc

-0.22

actics

-0.21

rophy

-0.20

İĺìĿ´

-0.20

.fromRGBO

-0.20

ynet

-0.20

ÑĤÐ²ÐµÑĢÐ¶Ð´Ð°

-0.19

POSITIVE LOGITS

wide

0.27

-wide

0.27

/world

0.26

-flag

0.25

-American

0.25

-made

0.24

-Israel

0.23

ä¹ĥèĩ³

0.22

èĮĥåĽ´åĨħ

0.21

');</

0.21

Activations Density 2.897%