INDEX

Explanations

baseline measurements or time points

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

当前

-1.50

is

-1.37

-1.31

 tanti

-1.30

 resa

-1.27

 neer

-1.24

 запах

-1.22

-1.20

no

-1.20

 blandt

-1.20

POSITIVE LOGITS

Život

1.59

 brachten

1.52

 чемпи

1.41

 leurs

1.40

öt

1.39

 same

1.38

ༀ

1.34

batsman

1.33

 arī

1.32

ја

1.30

Activations Density 0.157%