INDEX

Explanations

know more about

This neuron strongly activates on the phrase pattern “know more about” (and close variants) indicating an informational prompt.

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

 diesen

-1.66

 těchto

-1.58

 этих

-1.48

 estos

-1.47

ではなく

-1.46

 diese

-1.40

 може

-1.34

這樣的

-1.33

 otros

-1.30

 võib

-1.30

POSITIVE LOGITS

 about

1.80

why

1.73

檍

1.70

鮃

1.68

︕

1.55

 which

1.52

 whether

1.50

 validamos

1.40

dukan

1.39

⾞

1.36

Activations Density 0.029%