INDEX

Explanations

quoted phrases or words

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

 trusty

-1.18

ρό

-1.16

 zwanzig

-1.15

잤

-1.14

 chúng

-1.14

twenty

-1.13

 COULD

-1.12

 mehreren

-1.12

 miejsco

-1.11

 freakin

-1.10

POSITIVE LOGITS

 other

1.43

 also

1.43

 first

1.36

new

1.36

nienie

1.34

1.29

opheles

1.27

írus

1.27

("")

1.21

er

1.20

Activations Density 0.083%