INDEX

Explanations

holding or associated with

introduction or abstract drafts

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

 Lune

0.46

 Thunder

0.45

*"

0.42

 Swamp

0.42

Thunder

0.42

VEN

0.42

 Pale

0.41

 preponderance

0.41

 karta

0.41

Vu

0.41

POSITIVE LOGITS

DataMat

0.53

یت

0.52

 সার্ভিস

0.51

وینت

0.51

م

0.49

ম

0.48

یر

0.47

 الرس

0.47

 bénéficier

0.47

 обеспечи

0.46

Activations Density 0.000%