INDEX

Explanations

lists of purpose

common affirmative responses

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

Þ

0.51

 Gestión

0.49

Rite

0.49

愆

0.48

 Musica

0.47

נת

0.47

 કેવી

0.47

 ഭാര്യ

0.46

 posteriormente

0.46

 letzt

0.46

POSITIVE LOGITS

 rambling

0.48

 études

0.46

arms

0.45

indexing

0.45

সম

0.44

logical

0.44

arı

0.44

 illnesses

0.44

lists

0.43

ologische

0.43

Activations Density 0.000%