INDEX

Explanations

rx

This neuron fires on technical document section headings or API‐style labels—single capitalized terms (often followed by a period) like “module,” “function,” “UI,” “Payment Method,” “File,” etc.

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

ራል

0.78

ራሉ

0.70

 व्हाउचर

0.67

वारीत

0.67

 पहिल्या

0.66

èvres

0.64

 وكل

0.64

ებისა

0.63

 असू

0.63

 Boul

0.62

POSITIVE LOGITS

._

1.52

1.41

.__

1.39

().

1.23

.$

1.19

.___

1.13

。

1.13

.$.

1.11

_.

1.10

.{

1.06

Activations Density 0.140%