INDEX

Explanations

numbers and dates

The neuron detects four-digit numeric tokens corresponding to year values in dates.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

if

-1.93

 then

-1.86

 after

-1.83

and

-1.76

 should

-1.74

for

-1.74

 because

-1.59

 using

-1.57

 sonra

-1.55

も

-1.48

POSITIVE LOGITS

古怪

1.71

神情

1.70

靦

1.64

のだが

1.45

 désigner

1.44

 encontrada

1.43

 zweier

1.41

 shimmering

1.40

 않았

1.39

鸶

1.39

Activations Density 0.001%