INDEX

Explanations

yes or no questions

The neuron fires on floating‐point numeric tokens (numbers with decimal points) often used for specifications or measurements.

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

 fashioned

0.95

brochen

0.86

govt

0.81

phil

0.79

[,,"

0.78

kip

0.78

hamilan

0.77

google

0.76

fashioned

0.75

髭

0.74

POSITIVE LOGITS

ու

0.79

dž

0.74

 당시

0.72

 capables

0.71

 cultivés

0.71

씨

0.70

େ

0.70

 마치

0.69

 ガス

0.69

 organiques

0.68

Activations Density 0.001%