INDEX

Explanations

variable assignment `=`

The neuron fires on isolated single‐letter identifiers (e.g. “m”, “p”, “f”, “i”, etc.) in code.

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

 Schluß

-1.38

 Ayrıca

-1.26

なんですね

-1.24

際には

-1.23

んですね

-1.22

 Når

-1.22

為に

-1.22

skoda

-1.21

peugeot

-1.21

 dáms

-1.20

POSITIVE LOGITS

was

1.36

 that

1.25

is

1.25

は

1.23

1.21

new

1.21

1.14

搅拌

1.09

vocable

1.02

derry

1.02

Activations Density 0.001%