INDEX

Explanations

goes directly against principle

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

йте

1.21

LE

1.10

ّد

1.07

ृत

1.05

أت

1.04

 ridicul

1.03

 reliably

1.01

fot

0.99

چار

0.98

 someplace

0.98

POSITIVE LOGITS

MACHINE

1.39

ה

1.31

tı

1.31

ੀ

1.31

dzie

1.28

 disponibilidade

1.27

 забезпечення

1.27

 añadido

1.26

 değerlend

1.25

 masque

1.25

Activations Density 0.060%