INDEX

Explanations

TABLE expression

The neuron fires on isolated alphanumeric tokens—short uppercase words, single‐letter labels, table or section headings, and standalone numbers—rather than normal running text.

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

 there

-1.64

 usarlo

-1.36

 cómoda

-1.34

 последнее

-1.34

 ayudarte

-1.30

it

-1.30

 adecuada

-1.27

真实的

-1.27

 secar

-1.24

良かったです

-1.23

POSITIVE LOGITS

of

1.93

 with

1.69

化学

1.52

1.48

 być

1.46

,…

1.44

,\\

1.43

for

1.42

 perguntou

1.41

 have

1.41

Activations Density 0.007%