INDEX

Explanations

names and locations

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

ak

-0.65

akha

-0.64

 Kong

-0.59

sue

-0.54

 KONG

-0.52

athed

-0.52

enoord

-0.49

ipen

-0.49

 decoradas

-0.48

ing

-0.48

POSITIVE LOGITS

 itſelf

1.01

tagHelperRunner

0.94

 Monfieur

0.94

rungsseite

0.92

bootstrapcdn

0.92

 ―――――

0.90

 pleaſure

0.90

 Reſ

0.88

 Chriſt

0.86

 Jefus

0.84

Activations Density 0.062%