INDEX

Explanations

metadata key-value pairs

This neuron detects metadata field names or keys—uppercase or title‐case identifiers like “Map Name,” “Index,” “Default BGM,” “Version,” “Date,” “Msg id,” etc.

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

any

-1.44

 with

-1.38

 this

-1.30

if

-1.29

 believes

-1.28

at

-1.26

 from

-1.20

 blatantly

-1.19

",

-1.19

you

-1.17

POSITIVE LOGITS

Estab

1.30

données

1.26

produit

1.22

 utilizada

1.20

 especializada

1.20

？』

1.20

はなく

1.19

 complejos

1.15

Utili

1.15

 debilit

1.15

Activations Density 0.032%