INDEX

Explanations

date and time formats

This neuron activates on numeric tokens—especially multi‐digit numbers and decimals (measurements, timestamps, data values).

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

 Mileage

0.77

 Ouais

0.76

 milkshake

0.75

 PACE

0.71

 PFAS

0.71

MiG

0.71

 Horticulture

0.70

ʼ

0.70

 Albums

0.69

POSITIVE LOGITS

sql

0.72

一日

0.72

txt

0.67

Firefox

0.65

sunday

0.65

else

0.64

下

0.63

dateTime

0.63

slot

0.62

item

0.61

Activations Density 0.069%