INDEX

Explanations

The neuron specializes in spotting references to data resources—terms like “index,” “database,” “directory,” “provides,” etc., that describe online or published data collections.

database and archive content

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

but

-1.81

 three

-1.71

 four

-1.50

 five

-1.44

two

-1.38

 tuvieron

-1.27

 before

-1.26

 Seven

-1.25

 during

-1.25

 water

-1.25

POSITIVE LOGITS

 database

1.59

 GRATU

1.39

 corrig

1.37

 asma

1.35

 Motos

1.34

 kombat

1.30

 repository

1.30

 nunc

1.28

 Liens

1.27

 guitare

1.27

Activations Density 0.143%