INDEX

Explanations

since [date]

The neuron is triggered by four-digit year mentions (especially in “Since 19xx,” constructions).

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

什么呢

-1.77

儼

-1.74

瀟

-1.63

類型

-1.57

韪

-1.55

 nulle

-1.51

的時候

-1.50

 miała

-1.49

 ushered

-1.49

 choose

-1.48

POSITIVE LOGITS

’

2.38

.”

1.93

but

1.80

互相

1.73

 jedoch

1.70

1.69

:}

1.65

gemeinden

1.61

,{

1.58

1.52

Activations Density 0.051%