INDEX

Explanations

Gorbachev, hypotheses, slaves

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

 vendeur

2.58

 دستور

2.54

সঙ্গে

2.51

ތ

2.49

 tenido

2.44

atrième

2.43

 cabel

2.39

leine

2.37

aneous

2.35

 dode

2.34

POSITIVE LOGITS

stick

2.74

dig

2.64

σκο

2.63

am

2.62

 Initializes

2.59

chop

2.47

েরিক

2.44

ットフォーム

2.44

 trường

2.43

صة

2.42

Activations Density 0.002%