INDEX

Explanations

abstract concepts and states

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ADT

1.10

 cambi

0.88

ಂಡ

0.85

RULES

0.84

 الجنوب

0.84

ack

0.84

 reaping

0.84

 بير

0.84

IDEA

0.84

 накопи

0.83

POSITIVE LOGITS

 Excellency

1.81

 Highness

1.60

 Holiness

1.41

 Worship

1.32

 Emin

1.32

 excell

1.31

 Lordships

1.28

 lordship

1.27

 Majesty

1.22

 Honour

1.21

Activations Density 0.055%