INDEX

Explanations

contrasting information

The neuron activates on numeric tokens (integers and decimal 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

 only

-1.77

</u>

-1.71

},

-1.70

 during

-1.68

or

-1.66

 just

-1.60

 tikai

-1.56

";

-1.55

can

-1.53

 ļoti

-1.52

POSITIVE LOGITS

殽

2.33

鸲

1.93

媄

1.86

</h2>

1.67

⮑

1.62

丏

1.62

 efekt

1.54

 ellas

1.50

ouncing

1.49

 ความ

1.49

Activations Density 0.054%