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

/Library

-0.28

fade

-0.25

ework

-0.25

è¯·æĤ¨

-0.24

upil

-0.24

LIC

-0.24

 FALL

-0.23

elle

-0.23

æľīåĬ©

-0.23

è°ĥ

-0.23

POSITIVE LOGITS

ickets

0.27

 viá»ĩn

0.27

iska

0.26

 univers

0.26

éº»çĹ¹

0.26

 Pont

0.26

otten

0.25

atisfied

0.25

izens

0.25

è¿ĩå¤§

0.24

Activations Density 0.001%

No Known Activations

This feature has no known activations.