INDEX

Explanations

this:

The neuron fires on leading discourse markers that introduce an example or explanation—phrases like “goes like this:,” “here is the situation:,” etc.

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

搋

-1.31

 what

-1.30

érêt

-1.29

and

-1.28

能够

-1.25

 Senat

-1.22

蔬

-1.22

its

-1.22

Episodes

-1.20

蒜

-1.20

POSITIVE LOGITS

1.80

.”

1.56

 papai

1.43

 behaup

1.35

.</

1.34

 estadísticas

1.34

HostException

1.32

 estrategias

1.31

 fragte

1.30

odils

1.29

Activations Density 0.144%