INDEX

Explanations

abstract qualities or alternatives

The neuron flags formal, abstract legal or regulatory terms—especially key nouns and adjectives used in statutory definitions (e.g. “adverse,” “impact,” “quality,” “characteristic,” “management,” “policies,” etc.).

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

 একদম

0.52

éseket

0.47

 супер

0.45

ßerdem

0.44

shareButton

0.44

 breakpoints

0.44

餃

0.43

γρά

0.43

 зар

0.43

üllt

0.43

POSITIVE LOGITS

of

0.63

하거나

0.63

or

0.59

 veya

0.59

 eller

0.58

 oder

0.58

あるいは

0.57

または

0.56

 או

0.55

 thereof

0.55

Activations Density 0.000%