INDEX

Explanations

further change or effect

The neuron activates on verbs that express intensification or enhancement—especially ones meaning “improve,” “strengthen,” or “enhance.”

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

icestershire

-1.45

 aussi

-1.41

 גם

-1.40

 επίσης

-1.38

上げて

-1.38

llac

-1.37

 OGSÅ

-1.35

 también

-1.31

 նաև

-1.31

 différente

-1.30

POSITIVE LOGITS

 quando

1.48

 which

1.36

1.34

 bättre

1.34

the

1.33

1.30

\#:

1.27

 möjligt

1.27

 ketika

1.26

 because

1.25

Activations Density 0.021%