INDEX

Explanations

completion of action or state

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

 नवीन

1.21

 ১১

1.19

 Пер

1.18



1.18

 чай

1.18

 Ме

1.17

 Ди

1.17

 yoga

1.16

 ১৪

1.16

ვა

1.16

POSITIVE LOGITS

 pretty

1.43

 konkuren

1.33

 shitty

1.31

Plenty

1.28

 hordes

1.26

 kebanyakan

1.25

 cực

1.24

 ziemlich

1.24

 praticamente

1.23

 crappy

1.21

Activations Density 0.561%