INDEX

Explanations

decades ending in 's'

yields

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

igning

0.54

ğine

0.52

身的

0.52

‍♂️

0.50

omeness

0.50

anced

0.45

autical

0.44

ণ্ডের

0.44

likle

0.44

şu

0.43

POSITIVE LOGITS

ים

1.52

ों

1.49

們

1.46

们

1.37

‌های

1.34

وں

1.33

‌ها

1.33

1.27

‌هایی

1.24

கள்

1.23

Activations Density 0.139%