INDEX

Explanations

topics about soldiers, universe, or internet

The neuron primarily detects numeric tokens (e.g. numbers or numerical values) in the text.

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

CONCLUS

-1.50

硖

-1.49

 всех

-1.45

 alltid

-1.41

談社

-1.40

蚨

-1.39

 needy

-1.39

л

-1.38



-1.38

н

-1.37

POSITIVE LOGITS

and

1.79

 there

1.48

探し

1.36

 bezig

1.33

1.31

OREM

1.29

as

1.29

 makkelijk

1.28

 زیرا

1.27

1.23

Activations Density 0.236%