INDEX

Explanations

creative followed by specific terms

The neuron is highly activated by occurrences of the word “Creative.”

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

-3.52

 listos

-2.41

 praktisch

-2.31

am

-2.25

\"

-2.14

心中

-2.09

 involucrados

-2.05

籏

-2.03

\'

-2.02

POSITIVE LOGITS

that

3.33

 که

3.05

 That

2.63

颙

2.56

鋅

2.53

 phenomen

2.47

”،

2.47

姈

2.36

龊

2.34

2.28

Activations Density 0.005%