INDEX

Explanations

:**

The neuron detects numeric literals, particularly floating‐point numbers (i.e. tokens containing digits with decimal points).

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

 people

0.97

這個

0.95

我

0.93

this

0.93

这个

0.91

 this

0.90

我觉得

0.90

 मेरे

0.88

 मुझे

0.87

the

0.86

POSITIVE LOGITS

 meliputi

0.95

 diantaranya

0.78

 ofthe

0.73

 আসনের

0.71

 tartoz

0.71

<unused28>

0.70

含む

0.70

 расположен

0.68

🛋

0.68

(${

0.66

Activations Density 0.023%