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

ipel

-0.27

ipe

-0.26

matter

-0.26

æŀģåħ·

-0.25

trib

-0.24

ä¸ĢæľŁ

-0.24

heit

-0.23

_custom

-0.23

_CUSTOM

-0.23

åį´åıĪ

-0.22

POSITIVE LOGITS

infeld

0.30

 foreground

0.27

ITU

0.25

/she

0.24

++]

0.24

 cross

0.23

 ÑĢÐ¾Ð¶Ð´ÐµÐ½Ð¸Ñı

0.23

 visits

0.22

 markets

0.22

á»ĵ

0.22

Activations Density 0.887%

No Known Activations

This feature has no known activations.