INDEX

Explanations

time or duration

The neuron fires on numeric expressions—especially time intervals (e.g. “13 years,” “52 days,” “3 months,” etc.) and other quantified values.

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

inWeight

-1.65

outWeight

-1.64

droje

-1.46

rovna

-1.45

;&

-1.44

amd

-1.42

汌

-1.41

ﭽ

-1.32

(":");

-1.31

entación

-1.30

POSITIVE LOGITS

 there

1.66

it

1.61

there

1.60

 There

1.50

 tersebut

1.41

There

1.38

 when

1.35

 diversas

1.29

you

1.27

 various

1.24

Activations Density 0.069%