INDEX

Explanations

dates and months

The neuron detects date components (month abbreviations, day numbers, and years) 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

 ﬁrst

-1.38



-1.35

 แต่

-1.31

saat

-1.29

 arrend

-1.28

 universite

-1.25

}:

-1.24

Daftar

-1.23

пія

-1.23

皛

-1.23

POSITIVE LOGITS

to

1.65

可不是

1.52

 Figs

1.38

 aiutare

1.33

Its

1.32

 zespołu

1.30

by

1.29

 progettazione

1.29

OREM

1.29

 vrijdag

1.28

Activations Density 0.097%