INDEX

Explanations

type variable, in CCL, Houston, component

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

 даже

-1.36

 навіть

-1.30

-1.24

┍

-1.24

 nawet

-1.23

it

-1.17

setVertical

-1.16

 karena

-1.11

 médecin

-1.04

 اكتوبر

-1.03

POSITIVE LOGITS

 volledig

1.70

比べると

1.55

who

1.41

 köln

1.34

 confirmación

1.33

 especialidad

1.33

 prachtige

1.31

 gewi

1.31

 that

1.30

or

1.27

Activations Density 0.329%