INDEX

Explanations

sets of items

The neuron fires on standalone numeric tokens—quantities, measurements, prices, or other numbers—scattered through product descriptions.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.40

 الدولة

-1.27

 déplacer

-1.24

 what

-1.23

 zrobić

-1.21

だな

-1.20

籏

-1.20

 آنچه

-1.19

quée

-1.18

eaway

-1.18

POSITIVE LOGITS

 these

1.55

 this

1.34

our

1.31

 them

1.24

的名字

1.13

Cuando

1.13

、「

1.11

我們的

1.09

Antworten

1.06

//---

1.05

Activations Density 0.032%