INDEX

Explanations

used information or play

This neuron detects mentions of interactive or utilitarian interface elements (forms, links, apps, devices) and related instructions for using them.

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

 using

1.05

using

0.94

 Using

0.84

 usando

0.82

Using

0.80

 menggunakan

0.79

 utilizando

0.79

 USING

0.77

 utilizzando

0.77

USING

0.68

POSITIVE LOGITS

 sparingly

1.25

 wisely

0.96

 advantage

0.89

 interchangeably

0.88

instead

0.88

 extensively

0.87

 liberally

0.86

 hyvä

0.84

 instead

0.83

advantage

0.81

Activations Density 0.386%