INDEX

Explanations

code snippets and documentation formatting

The neuron is primarily activated by numeric tokens (i.e. standalone numbers or digit sequences).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 nominally

0.37

 renderEncoder

0.35

 กับ

0.35

 または

0.35

 그러나

0.34

літы

0.34

 lekin

0.33

 及び

0.33

 toolPath

0.33

 Daarnaast

0.33

POSITIVE LOGITS

 मेरी

0.40

আমার

0.38

 छात्र

0.37

मुझे

0.36

 विशाल

0.35

 огром

0.35

esthetic

0.35

 विद्यार्थी

0.35

accueil

0.35

 menonton

0.35

Activations Density 0.003%