INDEX

Explanations

The neuron flags occurrences of “icon” (and related forms like “icons” or “iconography”) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 effluents

0.60

 HttpResponse

0.57

 moelle

0.57

 catchment

0.56

 maturing

0.55

commonByteArray

0.55

 PrintWriter

0.55

 quadrupole

0.54

 récit

0.54

 CENTRAL

0.53

POSITIVE LOGITS

 icon

2.19

 icons

2.11

Icon

2.08

 Icon

2.00

 Icons

1.96

icon

1.92

Icons

1.87

图标

1.85

icons

1.81

アイコン

1.78

Activations Density 0.118%