INDEX

Explanations

hexadecimal numbers and colors

The neuron activates on sequences of hexadecimal digits (e.g. hex color codes or 0x… literals).

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

瑧

-1.83

 兔子

-1.72

<0xDE>

-1.65

 cláus

-1.64

 patří

-1.64

bigli

-1.60

 nech

-1.57

澌

-1.55

Lettuce

-1.55

 Cantidad

-1.55

POSITIVE LOGITS

You

2.00

）-

1.88

</h2>

1.85

 other

1.66

1.60

"){

1.55

...",

1.55

1.53

th

1.51

……」

1.45

Activations Density 0.040%