INDEX

Explanations

Norman origin, King John, eleventh century

The neuron primarily detects standalone numeric tokens (e.g., years, dates, and other numbers).

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

岡山

-1.35

静岡

-1.23

larımız

-1.20

🔻

-1.20

 than

-1.16

栃木

-1.16

 нам

-1.16

Having

-1.14

 آن‌ها

-1.13

syk

-1.12

POSITIVE LOGITS

 давления

1.30

olingo

1.27

 jazdy

1.27

 sukces

1.21

 chyba

1.20

 pokoj

1.17

 SPEC

1.16

 obyvateľov

1.16

 dosta

1.13

ванні

1.13

Activations Density 0.046%