INDEX

Explanations

common function words

The neuron activates on numeric tokens (especially decimal numbers), i.e. it finds numerical measurements in the text.

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

 first

-1.52

-1.48

と思っていた

-1.45

-1.40

-1.38

って思

-1.38

–

-1.37

und

-1.34

-1.30

 hijack

-1.30

POSITIVE LOGITS

kehren

1.57

になれば

1.52

 pudieron

1.47

шні

1.45

ských

1.44

 zelfs

1.42

銩

1.40

 Senin

1.38

 opport

1.37

ктория

1.37

Activations Density 0.298%