INDEX

Explanations

weather and forecasts

The neuron fires strongly on occurrences of the word “weather” (e.g. in product titles and headings).

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

 moonlight

0.62

 deja

0.59

 wavenumber

0.59

 evident

0.55

 seawater

0.55

 lighter

0.54

 wavefront

0.53

ा

0.53

 lights

0.53

 증가

0.53

POSITIVE LOGITS

Weather

1.09

weather

1.08

 WEATHER

1.04

 weather

1.00

⛈

1.00

 inclement

0.95

WEATHER

0.95

 Weather

0.93

 getWeather

0.91

 मौसम

0.89

Activations Density 0.195%