INDEX

Explanations

time ago

The neuron is a detector for numeric year or decade expressions (references to years such as “1970s,” “2012,” “two years ago,” etc.).

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

}";

0.57

 през

0.57

י

0.57

 בש

0.56

 outstretched

0.56

})

0.55

 inférieure

0.54

יא

0.54

ي

0.54

POSITIVE LOGITS

人们

0.80

但是

0.75

लेकिन

0.73

人們

0.73

फॉर्मेशन

0.73

 सामाजिक

0.71

যখন

0.70

它可以

0.69

行业的

0.68

 Многие

0.67

Activations Density 0.029%