INDEX

Explanations

date and time formats

The neuron activates on date/time format specifiers and their separators (e.g. %Y-%m-%d, HH:mm:ss) in code.

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

in

-1.81

gradable

-1.30

桄

-1.28

-1.26

but

-1.19

detek

-1.19

 nyel

-1.18

 münchen

-1.17

columnNumber

-1.17

<>

-1.16

POSITIVE LOGITS

不僅

1.29

 zumindest

1.28

 wenigstens

1.27

各種

1.27

どのような

1.25

 angefangen

1.25

事です

1.23

幾乎

1.21

 '/';

1.21

 decenas

1.19

Activations Density 0.005%