INDEX

Explanations

legal, medical, statistical, economic advice

The neuron strongly activates on occurrences of “Santa” in proper names (e.g. place or institution names beginning with “Santa”).

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

luoromethyl

0.55

䣬

0.54

geqslant

0.52

丙基

0.52

芬

0.52

越大

0.50

dem

0.50

espec

0.50

予

0.49

棘

0.49

POSITIVE LOGITS

्योर

0.68

 Clinic

0.64

 postdoc

0.63

 DEPARTMENT

0.62

 MARIA

0.62

 jogar

0.62

Maria

0.62

 DEVICE

0.62

 CONSULT

0.62

 hocam

0.61

Activations Density 0.034%