INDEX

Explanations

rectifying

The neuron strongly activates on the word stem “recti-” (as in “rectifier” or “rectifying”) and closely related DC power‐electronics terms.

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

 niños

-0.73

encanaan

-0.70

 evolución

-0.67

Գ

-0.66

 ľah

-0.66

óleo

-0.66

🎺

-0.66

sommer

-0.66

 speranza

-0.65

getChild

-0.63

POSITIVE LOGITS

 rectification

2.69

 recti

2.55

 rectifier

2.47

 rectified

2.36

 rectify

2.23

 RECT

1.95

RECT

1.87

Rec

1.70

 rect

1.66

 Rect

1.58

Activations Density 0.050%