INDEX

Explanations

column followed by name or header

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

ขา

1.87

脲

1.84

鵠

1.81

yuk

1.80

wenden

1.79

ружи

1.75

ARP

1.73

duke

1.73

distance

1.73

ema

1.71

POSITIVE LOGITS

 agrí

2.15

edinte

2.11

onial

2.05

 použit

2.04

ið

2.04

midrule

2.03

UMBIA

2.01

անի

1.97

osseum

1.97

olica

1.90

Activations Density 0.044%