INDEX

Explanations

rows affected

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tanks

-0.82

Ơ

-0.79

 bArr

-0.75

centiles

-0.75

๖

-0.75

 Acquisition

-0.74

LLocation

-0.74

ơ

-0.74

 スカート

-0.74

 excus

-0.74

POSITIVE LOGITS

 affected

1.48

 rowCount

1.48

 rows

1.46

Affected

1.46

affected

1.34

 Affected

1.27

row

1.27

rows

1.26

num

1.22

num

1.19

Activations Density 0.006%