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

vere

-0.31

 disease

-0.26

ALLOC

-0.26

&q

-0.25

å¯¹å¥¹

-0.25

vero

-0.25

 JACK

-0.25

Cot

-0.24

 emanc

-0.24

æľĪä¸ĭ

-0.24

POSITIVE LOGITS

fÃ¤hig

0.27

éĻĦè¿ĳçļĦ

0.27

åķ¡

0.26

è¾Ĩè½¦

0.26

/header

0.26

_NM

0.25

å¼ĢæĶ¾å¼ı

0.25

åħ°

0.25

æĹģè¾¹çļĦ

0.24

colm

0.24

Activations Density 0.000%

No Known Activations

This feature has no known activations.