INDEX

Explanations

technical definitions or descriptions

The neuron fires on lines of technical documentation formatting—ASCII‐style separators, headers, URLs, and other code‐block markup.

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

 like

-1.45

 then

-1.27

:^{

-1.26

-1.24

 Durchschnitt

-1.20

 accesorios

-1.19

 świet

-1.17

 życia

-1.17

so

-1.17

POSITIVE LOGITS

 Эти

1.70

丹药

1.68

嵎

1.45

亟

1.42

ῖν

1.38

ὴν

1.38

znym

1.38

 chirurgie

1.37

sieht

1.37

gewöhn

1.37

Activations Density 0.005%