INDEX

Explanations

days and months

The neuron is tuned to detect date expressions—particularly mentions of months, days of week, and numerical day/year components.

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

 görmek

-1.74

 должны

-1.64

对于

-1.59

 musíte

-1.57

 destinés

-1.55

 digelar

-1.53

お肉

-1.53

擽

-1.51

 Weisheit

-1.50

 possano

-1.48

POSITIVE LOGITS

in

1.79

 with

1.71

on

1.66

my

1.57

his

1.53

 junt

1.49

had

1.48

and

1.47

1.41

 while

1.41

Activations Density 0.026%