INDEX

Explanations

consistency and collection

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

 rescind

1.54

 primul

1.44

 तत्काल

1.43

 shears

1.39

 उर्फ

1.37

 जबकि

1.35

ंना

1.34

LAC

1.33

 viewHolder

1.32

sue

1.30

POSITIVE LOGITS

\">

1.29

…</

1.27

 يد

1.26

اقة

1.26

مة

1.26

 خبر

1.24

 الأرض

1.23

د

1.21

ن

1.21

 تد

1.21

Activations Density 0.000%