INDEX

Explanations

court action and proceeding identifiers

The neuron is picking up on legal case identifiers and docket‐style numeric citations (numbers, decimals, and “No.” labels).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 portugu

-1.56

 same

-1.48

oprecipitation

-1.47

zystać

-1.41

</h2>

-1.41

of

-1.36

 авто

-1.36

鯖

-1.34

use

-1.30

各大

-1.30

POSITIVE LOGITS

momile

1.48

꜆

1.48

 impair

1.45

 nastro

1.40

zzie

1.37

 необхідно

1.37

 hermoso

1.35

猁

1.34

ſtion

1.34

you

1.33

Activations Density 0.174%