INDEX

Explanations

derivative of single variables

The neuron fires on numeric tokens (especially multi‐digit or decimal numbers) in the text.

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

 altre

-1.30

if

-1.30

하면

-1.19

 maisons

-1.19

 then

-1.18

 momente

-1.17

其他的

-1.16

Then

-1.12

-1.11

也有

-1.09

POSITIVE LOGITS

がありません

1.38

 zprávy

1.29

erkek

1.27

昵

1.26

 една

1.25

wnież

1.25

 souligné

1.24

 délai

1.20

 Réponses

1.20

 impresiones

1.20

Activations Density 0.059%