INDEX

Explanations

describing voice quality

This neuron responds to vivid narrative event markers—especially character names, descriptive sensory details, and action verbs that signal key story moments.

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

私が

0.64

 escorted

0.61

私は

0.55

я

0.54

 אני

0.54

₹

0.53

 crad

0.52

",{

0.51

 turntable

0.50

让我

0.50

POSITIVE LOGITS

 말했다

0.74

 분석

0.67

 નિર્ણય

0.66

 análise

0.66

 মনোভাব

0.64

 সমী

0.63

不禁

0.63

明白

0.62

 analisis

0.62

 analiza

0.62

Activations Density 0.018%