INDEX

Explanations

Scott Walker, option

The neuron fires on programming‐ or data‐format identifiers and metadata tokens—e.g. library/class names (defaultdict), XML tags (TS, version), and version number strings.

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

 TARGET

-0.84

target

-0.80

raca

-0.78

 ovarian

-0.77

ضافة

-0.77

free

-0.72

TARGET

-0.71

🤶

-0.71

 塗装

-0.70

 Sheng

-0.69

POSITIVE LOGITS

 Devlin

0.71

 BOLD

0.71

Muh

0.68

Gez

0.67

天井

0.65

bukti

0.64

*>(

0.64

 Enquiry

0.63

ọn

0.61

Rispondi

0.61

Activations Density 0.070%