INDEX

Explanations

fine, A, ft. measurement or clarification

The neuron activates on numeric tokens (stand-alone numbers, years, measurements, or other digit sequences).

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

 hoje

0.97

 oficialmente

0.95

 financiero

0.94

 میں

0.93

أن

0.92

กลาง

0.91

Segundo

0.91

 incrível

0.90

 މި

0.90

 Klopp

0.90

POSITIVE LOGITS

ы

0.78

м

0.77

 downward

0.73

 twofold

0.72

ౖ

0.69

 order

0.67

ı

0.67

 appeal

0.66

ро

0.65

 therein

0.63

Activations Density 0.000%