INDEX

Explanations

interpretation of styles

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

urin

0.49

Temp

0.47

jskiego

0.47

حی

0.45

Gruppe

0.44

illerie

0.43

是通过

0.43

jsko

0.42

𝑤

0.42

صر

0.42

POSITIVE LOGITS

 Interpretation

0.52

 interpret

0.51

 tolu

0.48

 Styles

0.47

thiophen

0.47

फेर

0.45

ए

0.45

 Sources

0.45

 Estimates

0.45

 Interpre

0.44

Activations Density 0.001%