INDEX

Explanations

plural nouns followed by list syntax

The neuron fires on identifiers used as object or class property names (e.g. “Products”, “Cells”, “childs”).

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

if

-1.59

in

-1.52

at

-1.43

on

-1.27

 although

-1.26

of

-1.24

 when

-1.23

 method

-1.20

 before

-1.20

 from

-1.20

POSITIVE LOGITS

 tormenta

1.51

ఝ

1.45

 such

1.43

 ANCHE

1.42

 будущего

1.41

']),

1.38

 associés

1.35

 granja

1.34

 playera

1.34

 bestia

1.30

Activations Density 0.045%