INDEX

Explanations

have a, have made, have been

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

or

-2.06

 which

-1.88

攵

-1.48

隹

-1.33

"")

-1.31

ñado

-1.20

 oraz

-1.19

who

-1.18

 której

-1.18

and

-1.16

POSITIVE LOGITS

 väldigt

1.45

1.22

非常に

1.18

 kinda

1.18

 достаточно

1.16

本当に

1.14

いろんな

1.13

 córdoba

1.13

 wani

1.13

 dégustation

1.12

Activations Density 0.013%