INDEX

Explanations

geographical locations

The neuron activates on numeric tokens—particularly numbers (including decimals) representing measurements, statistics, dates, etc.

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

 which

-1.29

but

-1.28

了

-1.21

 beant

-1.20

lon

-1.17

 human

-1.13

だが

-1.10

団

-1.08

 sendo

-1.06

 persones

-1.05

POSITIVE LOGITS

oth

1.23

 تھے

1.22

ांकि

1.21

ے

1.21

 europeos

1.21

{};

1.18

 jepang

1.13

 Tüm

1.12

鉀

1.09

 بھ

1.08

Activations Density 0.080%