INDEX

Explanations

thrown

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

 endorsements

-0.27

CHO

-0.26

è®·

-0.24

Cla

-0.24

âĭ®

-0.23

è¢«æīĵ

-0.23

ivor

-0.23

 clicking

-0.23

 verk

-0.23

ductive

-0.23

POSITIVE LOGITS

 bonus

0.32

bonus

0.31

Bonus

0.30

 extra

0.29

 kicks

0.29

é¢Ŀå¤ĸ

0.29

éĻĦåĬł

0.28

.extra

0.27

_bonus

0.27

 Bonus

0.27

Activations Density 0.010%