INDEX

Explanations

colors, sizes, compatible

The neuron is triggered by numeric specifications and measurements (e.g. dimensions, sizes, model numbers).

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

🪛

-1.55

🫓

-1.51

מונות

-1.51

 queijo

-1.46

 lokale

-1.40

 ludzie

-1.40

 couro

-1.35

 hermosas

-1.34

 několika

-1.34

⟼

-1.29

POSITIVE LOGITS

 with

1.78

 provides

1.37

𝗼

1.30

ꗃ

1.29

 used

1.25

";

1.25

 also

1.25

 demonstrates

1.23

 Abbé

1.21

 Mlle

1.21

Activations Density 0.091%