INDEX

Explanations

There

This neuron spotlights high-salience discourse markers—words and punctuation that signal questions, causal links, emphatic statements, or other rhetorical transitions.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

这里的

0.69

 aqui

0.63

 disini

0.62

 اینجا

0.61

 here

0.59

这一切

0.59

ils

0.58

сюда

0.58

HERE

0.57

 här

0.56

POSITIVE LOGITS

 there

5.06

there

4.63

 هناك

4.50

 THERE

4.09

There

3.98

 There

3.81

THERE

3.66

 theres

3.34

 وہاں

3.27

 वहां

3.23

Activations Density 1.371%