INDEX

Explanations

academic readings and chapters

summarizing articles and texts

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

qh

0.45

urbo

0.43

 rarest

0.41

 unscrupulous

0.41

 populated

0.41

 planned

0.40

 backlog

0.40

 dédié

0.40

 geometry

0.40

രിച്ചു

0.40

POSITIVE LOGITS

 excerpt

0.86

Excerpt

0.75

 excerpts

0.74

excerpt

0.64

 Chapter

0.57

 প্রবন্ধ

0.57

 Readings

0.57

 Kapitel

0.57

 readings

0.56

观看

0.55

Activations Density 0.034%