INDEX

Explanations

days ago or old

The neuron detects numeric expressions denoting time intervals (e.g. counts of days or weeks).

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

-1.95

eynman

-1.58

ٞ

-1.52

-1.47

子上

-1.44

荦

-1.40

入って

-1.38

 beinhaltet

-1.38

部分は

-1.37

 durante

-1.34

POSITIVE LOGITS

因为

1.70

我们的

1.63

賔

1.58

 soucis

1.55

こんばん

1.53

שלום

1.49

因為

1.48

Because

1.48

澳門

1.46

ETIES

1.46

Activations Density 0.044%