INDEX

Explanations

reliance on something

The neuron fires on words that express dependence or reliance (e.g., “relying,” “reliance,” “dependence”).

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

 geeign

0.57

 appropriate

0.56

{*}

0.54

 Hurley

0.54

เหมาะสม

0.53

 осторо

0.53

নমেন্ট

0.52

ካ

0.52

 нейтра

0.52

နိုင်

0.50

POSITIVE LOGITS

 rely

3.31

 relies

3.13

 reliance

3.13

 relying

3.06

依赖

2.95

 reliant

2.95

 Rely

2.91

依靠

2.89

 relied

2.83

reliance

2.81

Activations Density 0.643%