INDEX

Explanations

date formats like YYYY-MM-DD

The neuron activates on the “m” tokens used as month specifiers in date/time format strings.

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

.’

-2.25

 setelah

-1.88

 also

-1.88

]^{

-1.77

("")]

-1.74

after

-1.71

 после

-1.71

</h4>

-1.69

 иногда

-1.62

”

-1.55

POSITIVE LOGITS

夲

1.74

蟶

1.69

 ſte

1.68

 faſt

1.64

她們

1.58

 ſta

1.57

 HISTOR

1.56

锨

1.55

 sélections

1.54

 serisi

1.53

Activations Density 0.002%