INDEX

Explanations

assigning or relating causes

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

ты

-1.15

><

-1.10

 zarówno

-1.05

がこの

-1.05

 "/";

-1.03

 تلك

-1.03

cún

-1.02

นี้

-1.02

 diesem

-1.01

覚醒

-1.01

POSITIVE LOGITS

なんですね

1.38

any

1.34

 then

1.31

 anything

1.31

gelijke

1.25

 čier

1.20

 piatti

1.16

}(),

1.13

女生

1.12

 možnosti

1.11

Activations Density 0.091%