INDEX

Explanations

The main thing this neuron does is detect geographic place names (cities, states, provinces, etc.).

geographical locations

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

or

-1.66

in

-1.55

and

-1.46

so

-1.45

of

-1.38

 then

-1.29

 from

-1.29

if

-1.28

 like

-1.24

-1.22

POSITIVE LOGITS

 conster

1.41

 étranger

1.30

 capot

1.25

 perpé

1.23

 imprimé

1.21

 ondu

1.20

 marte

1.20

 vulga

1.20

 luto

1.20

 auront

1.19

Activations Density 0.024%