INDEX

Explanations

north and south regions

The neuron fires on cardinal‐direction terms and related regional descriptors (e.g. north, southern, northern, south).

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

<0xE1>

-1.49

órmulas

-1.49

 учетом

-1.38

 вона

-1.35

What

-1.33

一颗

-1.32

 kwietnia

-1.28

 récente

-1.27

 there

-1.26

ためには

-1.23

POSITIVE LOGITS

of

1.66

’

1.45

での

1.37

ですが

1.28

的朋友

1.27

などが

1.27

に合わせて

1.20

だけど

1.17

了她

1.17

 unités

1.16

Activations Density 0.016%