INDEX

Explanations

creative expression and self

The neuron fires on abstract, content-rich words in reflective/expository passages—i.e. terms expressing high-level concepts rather than concrete nouns or simple function words.

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

ּוֹ

-1.87

ִּ

-1.82

ַּ

-1.77

 اندروید

-1.62

ֶּ

-1.55

ֵּ

-1.54

ְּ

-1.52

 лекарства

-1.51

 موبایل

-1.45

 السياس

-1.44

POSITIVE LOGITS

ׇ

1.61

ὁ

1.49

ἀ

1.41

 myself

1.31

 عاشقانه

1.29

 Fungsi

1.25

 مجلسی

1.24

𝑆

1.23

 растений

1.22

 Sebuah

1.22

Activations Density 0.031%