INDEX

Explanations

Junior, Mid-Level, Senior

The neuron strongly activates on floating‐point numeric tokens (numbers containing a decimal point).

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

seu

0.61

ap

0.61

<0xC2>

0.59

 amigas

0.58

 problemas

0.58

 bigint

0.57

\\

0.57

 problema

0.57

 suas

0.57

 suficiente

0.57

POSITIVE LOGITS

accoon

0.72

 என்பவர்

0.72

 специалист

0.71

anchen

0.67

 مدير

0.66

 יל

0.66

 якого

0.64

 금융

0.63

 ビジネス

0.63

 Directeur

0.63

Activations Density 0.000%