INDEX

Explanations

workaround, community, rationale, exploits

The neuron primarily detects numeric expressions and quantities (e.g. multi‐digit numbers and statistics).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

Whilst

1.41

 nuestras

1.27

 Whilst

1.26

 наших

1.21

 VERY

1.20

 nuestra

1.19

 нашего

1.19

 НЕ

1.17

 нашей

1.16

١

1.16

POSITIVE LOGITS

 지난해

1.15

 मुहैया

1.07

 staffers

1.06

 yüzde

1.05

 amid

1.03

<unused2162>

0.98

 například

0.96

 apiece

0.94

<unused2169>

0.94

—

0.93

Activations Density 2.968%