INDEX

Explanations

with or although

The neuron activates specifically on numeric tokens and measurements (e.g. digits with or without decimals).

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

 його

-1.37

几次

-1.35

of

-1.27

donc

-1.24

 высокая

-1.24

immagine

-1.24

 suoi

-1.23

 singoli

-1.22

 resounding

-1.22

+}$

-1.22

POSITIVE LOGITS

 już

1.34

已经

1.31

we

1.28

 funcione

1.27

 While

1.26

 már

1.25

 Although

1.24

任何

1.23

 यह

1.22

 почти

1.22

Activations Density 0.156%