INDEX

Explanations

section titles and lists

The neuron detects numeric and quantitative information—numbers, dates, rankings, counts, and other statistics.

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

now

-1.53

 also

-1.34

 then

-1.33

Дереккөздер

-1.30

 требует

-1.28

</em>

-1.22

harmed

-1.20

或者是

-1.20

ètres

-1.18

etration

-1.17

POSITIVE LOGITS

 была

1.25

 ""));

1.19

Orléans

1.17

Život

1.16

canActivate

1.16

Unnamed

1.15

There

1.15

anglicky

1.13

 расчет

1.13

 siempre

1.12

Activations Density 0.344%