INDEX

Explanations

code identifiers

The neuron fires on hexadecimal digits (0–9, a–f) within long ID or GUID strings.

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

-1.44

 will

-1.41

as

-1.34

of

-1.30

are

-1.16

pení

-1.13

 econômica

-1.10

 your

-1.06

JUL

-1.05

 Ankunft

-1.05

POSITIVE LOGITS

 dégâts

1.59

 rác

1.37

 RESPONDENT

1.35

 télécharge

1.30

 Meningkatkan

1.30

 désol

1.30

 UserController

1.29

槩

1.29

 roligt

1.27

mbggenerated

1.26

Activations Density 0.212%