INDEX

Explanations

movement and direction

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

 such

-1.52

to

-1.39

 аналоги

-1.38

 vélos

-1.34

瘴

-1.32

now

-1.29

：……

-1.28

 саду

-1.25

ীয়

-1.25

他们

-1.23

POSITIVE LOGITS

1.59

=()=>{

1.45

 anunció

1.39

ᾳ

1.38

 тільки

1.35

 estaría

1.32

還會

1.30



1.28

你觉得

1.27

 revisar

1.26

Activations Density 0.207%