INDEX

Explanations

Generally, ideally, typically

The neuron is keyed to detecting adverbial qualifiers or hedging words (e.g. “generally,” “usually,” “automatically,” “typically,” “normally”) that modify or qualify statements.

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

বিভিন্ন

0.84

 необы

0.81

ต

0.80

मय

0.77

明显

0.76

ᓘ

0.76

溂

0.75

organisme

0.74

注意

0.74

अंतर

0.74

POSITIVE LOGITS

 default

0.93

 defaults

0.90

default

0.82

only

0.82

ade

0.80

ees

0.79

 quidem

0.78

0.77

 여기에

0.77

 aber

0.76

Activations Density 0.870%