INDEX

Explanations

based on preference or need

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

भूति

1.38

 entweder

1.38

 either

1.32

𝚒

1.30

1.28

HintPath

1.28

Iso

1.27

াইবার

1.27

\\

1.26

 justement

1.23

POSITIVE LOGITS

but

1.66

BUT

1.47

But

1.43

 pero

1.43

But

1.42

なのに

1.41

but

1.41

이지만

1.37

ですが

1.37

DA

1.33

Activations Density 0.068%