INDEX

Explanations

contrasting with or unlike

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

-1.15

̴

-1.10

 wünscht

-1.09

 rám

-1.06

 ugyan

-1.05

];

-1.02

and

-1.01

ก็

-1.00

も同じ

-0.99

🖐

-0.96

POSITIVE LOGITS

 here

1.28

 which

1.23

 this

1.23

 where

1.20

就没有

1.10

นี้

1.06

是没有

1.04

which

1.02

 doesn

1.02

 δεν

0.99

Activations Density 0.014%