INDEX

Explanations

inherent properties and qualities

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

</h1>

-2.95

 visse

-2.64

 dicono

-2.36

瓩

-2.34

 visar

-2.28

</i>

-2.25

sistence

-2.23

衕

-2.22

It

-2.19

 Jér

-2.17

POSITIVE LOGITS

ковой

2.61

ଽ

2.33

诸多

2.22

şi

2.19

 ול

2.19

 heath

2.17

 grappling

2.11

اً

2.11

 $^{\

2.08

ского

2.06

Activations Density 0.005%