INDEX

Explanations

hole

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hole

-2.38

hole

-2.00

 Hole

-1.95

Hole

-1.92

 holes

-1.75

 HOLE

-1.74

HOLE

-1.60

holes

-1.58

 Holes

-1.47

 agujero

-1.34

POSITIVE LOGITS

Demografie

0.60

principalColumn

0.59

UpdateCommand

0.59

 beginnetje

0.58

SequentialGroup

0.56

LayoutPanel

0.54

hares

0.53

 EnglishChoose

0.53

EndGlobalSection

0.53

 محفوظة

0.52

Activations Density 0.128%