INDEX

Explanations

form fields

The neuron fires on form fields or labels that contain or expect numeric values (e.g. phone numbers, zip/postal codes, dates, and other digit‐based input fields).

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

į

-1.33

tych

-1.27

rethe

-1.27

Į

-1.27

wouldn

-1.23

 edin

-1.22

albeit

-1.21

◾

-1.21

heh

-1.20

mäh

-1.17

POSITIVE LOGITS

if

1.97

 your

1.87

 discernible

1.34

できれば

1.29

最大の

1.23

or

1.23

 should

1.20

any

1.16

如果不

1.13

 cheerful

1.11

Activations Density 0.017%