INDEX

Explanations

specific fields and subjects

The neuron detects numeric quantities and measurements (especially decimals) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 lleno

-1.55

četně

-1.48

 bílá

-1.44

 dře

-1.35

 šed

-1.34

 angiel

-1.34

 modrá

-1.31

 intentando

-1.31

 bílé

-1.30

 cuyas

-1.30

POSITIVE LOGITS

1.65

as

1.50

1.38

 like

1.38

alties

1.31

 prostitu

1.28

 cerdas

1.27

 romantis

1.24

「

1.24

 logis

1.23

Activations Density 0.256%