INDEX

Explanations

date separators and months

The neuron predominantly activates on numeric tokens (e.g. years, ages, and other numbers) in the text.

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

 Thursday

0.82

 Friday

0.76

Thursday

0.74

 Tuesday

0.73

Friday

0.72

 jeudi

0.71

Tuesday

0.70

 Saturday

0.68

Saturday

0.67

 Wednesday

0.67

POSITIVE LOGITS

ប្រស

0.63

 Prussian

0.61

秦

0.52

 letzt

0.51

 वैलेंट

0.51

 Wied

0.50

 stein

0.50

 einfachen

0.50

 Königs

0.50

দর্শন

0.49

Activations Density 0.131%