INDEX

Explanations

code file paths

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

 poë

-1.35

 applau

-1.34

ïe

-1.32

翃

-1.31

 geografic

-1.27

гают

-1.22

ӓ

-1.20

 geograf

-1.17

dientemente

-1.17

 corriger

-1.17

POSITIVE LOGITS

 Initially

1.49

at

1.46

not

1.41

 pertanian

1.36

It

1.34

 Also

1.32

 ONLY

1.31

on

1.30

 pesawat

1.30

 radically

1.28

Activations Density 0.011%