INDEX

Explanations

potentiation / enhancement

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

al

-2.78

穑

-2.59

 sont

-2.39

 annuel

-2.23

基本的

-2.19

enablog

-2.16

 voor

-2.09

 intrigue

-2.09

 fateful

-2.06

外人

-2.03

POSITIVE LOGITS

鲧

3.05

釹

2.89

and

2.63

紲

2.45

if

2.42

 unangemess

2.34

ांकि

2.28

 dúvidas

2.28

而在

2.25

 Они

2.20

Activations Density 0.013%