INDEX

Explanations

other contexts

The neuron activates on numeric tokens (digits, numbers, and numeric 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

 zarówno

-1.38

はもちろん

-1.35

 both

-1.31

ğu

-1.30

 době

-1.30

 neither

-1.29

ITHER

-1.27

 either

-1.22

ścioła

-1.19

];

-1.16

POSITIVE LOGITS

 other

1.59

1.42

 andere

1.25

Other

1.22

1.17

ных

1.16

 otros

1.14

その他の

1.13

 helps

1.09

其他

1.09

Activations Density 0.087%