INDEX

Explanations

government minister titles

The neuron detects references to government agencies, ministries, committees, or official administrative bodies.

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

 другой

-1.48

 других

-1.45

oth

-1.40

 其他

-1.37

 étaient

-1.33

 они

-1.33

 них

-1.30

责编

-1.29

 hordes

-1.28

 autres

-1.27

POSITIVE LOGITS

 after

1.58

 also

1.56

and

1.55

at

1.55

by

1.53

an

1.51

},

1.51

{\

1.49

зец

1.48

 should

1.43

Activations Density 0.048%