INDEX

Explanations

code tokens like parentheses and operators and following 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

of

-1.33

out

-1.19

 decimos

-1.09

-1.06

="

-1.05

VERSION

-1.05

gamos

-1.05

 instanceof

-1.03

 CGRectMake

-1.02

を行う

-1.01

POSITIVE LOGITS

when

1.25

we

1.23

нави

1.20

 identifies

1.20

 επίσης

1.17

 også

1.16

 également

1.16

…’

1.13

 myös

1.13

いよいよ

1.12

Activations Density 0.032%