INDEX

Explanations

starting of phrases

The neuron fires on stray punctuation tokens—especially isolated dashes and empty‐parameter parentheses—that tend to appear in code or template fragments.

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

zestaw

-1.32

zdroj

-1.29

 öğretmen

-1.26

sprzed

-1.20

źródło

-1.17

昨日の

-1.16

 ücretsiz

-1.16

 ujedno

-1.16

zdjęcie

-1.16

 sitten

-1.16

POSITIVE LOGITS

などは

1.49

1.41

st

1.31

では

1.29

1.28

int

1.27

1.26

le

1.25

Activations Density 0.000%