INDEX

Explanations

numbers followed by units

The neuron detects numeric tokens and numeric measurements (digits, numbers, and units like mm, GB, MHz — i.e., quantities and model/version numbers).

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

<unused2221>

0.50

在

0.43

:}

0.40

 disobey

0.40

ಗೊಂಡ

0.40

 در

0.39

'.$

0.38

 dissent

0.38

 في

0.38

কে

0.37

POSITIVE LOGITS

iculture

0.41

ρα

0.40

ουμε

0.39

 क्षमता

0.38

いますが

0.38

ουλ

0.38

ﻄ

0.38

います

0.37

ﺼ

0.37

kách

0.37

Activations Density 1.039%