INDEX

Explanations

is followed by common words

various languages

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

qn

0.45

 herewith

0.43

clare

0.38

 whereof

0.37

 Roshan

0.37

*}{

0.36

[…]

0.36

 awhile

0.36

TPM

0.36

চক

0.36

POSITIVE LOGITS

يح

0.40

🏻

0.39

 зат

0.38

 erat

0.38

ъ

0.37

UTION

0.36

인데

0.35

詹

0.35

 হচ্ছিল

0.34

 দেয়া

0.34

Activations Density 0.020%