INDEX

Explanations

code and links

This neuron strongly activates on segments of Google Code project URLs (i.e. “code.google.com/p/…”).

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

分かり

-0.87

は見

-0.78

!<

-0.77

 Erkrankung

-0.76

 Восто

-0.76

多彩

-0.75

 robe

-0.75

 honourable

-0.74

大事

-0.73

\'{

-0.72

POSITIVE LOGITS

 Code

1.00

ونز

0.81

 који

0.80

DON

0.79

 panoramic

0.78

但是

0.77

 اخبار

0.77

 ممت

0.77

当地

0.77

Code

0.76

Activations Density 0.149%