INDEX

Explanations

Federal Circuit court cases

The neuron fires on institutional headers—especially references to the “United States Court of Appeals for the Federal Circuit.”

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

by

-1.32

of

-1.29

лно

-1.27

 Kedua

-1.25

 Lakukan

-1.24

ly

-1.23

цій

-1.18

 Capitán

-1.17

://"

-1.16

 Semoga

-1.16

POSITIVE LOGITS

 about

1.45

 tentang

1.45

<bos>

1.38

 pensado

1.36

 précédente

1.34

 intentos

1.34

 muertes

1.34

を入れ

1.33

 uwag

1.32

RITION

1.31

Activations Density 0.119%