INDEX

Explanations

month names in dates

The neuron detects calendar date expressions (especially “On <Month> <Day>, <Year>” and similar month/day/year patterns).

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

who

-1.49

to

-1.48

 remaining

-1.32

赤ちゃん

-1.30

 inclusion

-1.27

hmmm

-1.27

 including

-1.24

 initially

-1.23

 simply

-1.23

but

-1.22

POSITIVE LOGITS

of

1.99

riages

1.30

주의

1.30

ческого

1.29

悫

1.27

Tentang

1.27

氳

1.25

 vrienden

1.25

iosis

1.23

 kemarin

1.23

Activations Density 0.145%