INDEX

Explanations

company names

The neuron is keyed to tokens representing structural separators—especially hyphens and slashes—used in codes, compound names, and numeric identifiers.

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

 during

-1.43

and

-1.33

 because

-1.25

mOn

-1.23

か月

-1.16

</b>

-1.13

as

-1.13

[:,:,

-1.11

 change

-1.11

 also

-1.11

POSITIVE LOGITS

？』

1.35

Sabato

1.32

↵

1.27

 alguna

1.24

 einige

1.16

 אם

1.16

 年级

1.16

 vorne

1.16

 medes

1.16

 lät

1.15

Activations Density 0.130%