INDEX

Explanations

accepted, waived, revealed, wants

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

apo

-0.08

ellas

-0.08

arte

-0.08

eus

-0.08

ãĤ§

-0.08

-strokes

-0.07

cur

-0.07

indi

-0.07

-ap

-0.07

amar

-0.07

POSITIVE LOGITS

atten

0.11

 Freem

0.10

Ã³r

0.09

ited

0.09

 Seamless

0.09

 explan

0.09

 Voll

0.08

ex

0.08

nullptr

0.08

 nullptr

0.08

Activations Density 0.001%

No Known Activations