INDEX

Explanations

numbers and citations

43

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

ила

0.43

ড়ার

0.42

 ラウンド

0.41

閎

0.41

ilha

0.40

 পরিস্থিতি

0.40

 NAACP

0.39

urusan

0.39

ﮏ

0.39

uska

0.38

POSITIVE LOGITS

Geb

0.41

 gain

0.39

Kok

0.38

មាន

0.38

Gain

0.36

 Gain

0.35

Representation

0.35

Sec

0.34

 विक

0.34

 Kinney

0.34

Activations Density 0.008%