INDEX

Explanations

keywords followed by types

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

Ży

-1.25

Zü

-1.23

еру

-1.21

malte

-1.20

-1.19

～！

-1.18

ränk

-1.16

ancang

-1.16

 essencial

-1.15

 възможно

-1.15

POSITIVE LOGITS

譏

1.64

了一种

1.34

 alcune

1.32

ية

1.30

ance

1.29

pable

1.29

son

1.27

ia

1.26

to

1.26

了一個

1.24

Activations Density 0.123%