INDEX

Explanations

without exceptions or limitations

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

үй

2.68

 اړه

2.44

𝒐

2.35

ים

2.30

 ونص

2.25

asjonen

2.24

asjoner

2.24

umab

2.21

φή

2.21

POSITIVE LOGITS

ें

2.98

 nhàng

2.80

\]

2.69

mares

2.69

ful

2.58

ment

2.54

man

2.51

bulb

2.42

𝚒

2.42

ૃત

2.42

Activations Density 0.704%