INDEX

Explanations

to a level

The neuron detects when a text mentions a measured “minimum” value or the idea of something slowing down to its lowest point.

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

of

-1.86

 people

-1.77

 then

-1.28

 invigor

-1.27

 went

-1.27

 certainly

-1.27

 keeps

-1.27

 these

-1.25

or

-1.23

 spurred

-1.20

POSITIVE LOGITS

 quintessential

1.52

 REALLY

1.48

 something

1.47

 conhecida

1.46

 feita

1.43

 sooo

1.39

⟣

1.38

 somewhat

1.38



1.35

 soooo

1.34

Activations Density 0.161%