INDEX

Explanations

preserve, preservation, preserving, prescreen

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

瘘

1.83

Curs

1.75

стор

1.74

Cv

1.74

újt

1.72

ostomy

1.71

 znaj

1.67

вт

1.66

 моги

1.66

齐

1.65

POSITIVE LOGITS

apples

1.57

DEP

1.43

 대상

1.35

 Targets

1.33

ollip

1.33

 swoop

1.33

မြို့

1.32

ampang

1.30

 ameri

1.29

EO

1.28

Activations Density 0.010%