INDEX

Explanations

is

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

_OTHER

-0.29

åħ¶ä»ĸ

-0.29

 OTHER

-0.28

 autres

-0.28

Other

-0.28

æľ¬çĿĢ

-0.27

 outras

-0.27

 other

-0.27

(other

-0.27

é¡¶

-0.26

POSITIVE LOGITS

 ones

0.32

 advantageous

0.26

 vice

0.25

DAR

0.24

predictions

0.24

 predictions

0.23

__()↵↵

0.23

 Bruins

0.23

StackTrace

0.23

ues

0.23

Activations Density 0.006%