INDEX

Explanations

times and quantities

This neuron activates on numeric measurement tokens—especially decimal or statistical figures (e.g. yardage, averages, percentages).

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

not

-2.16

Not

-1.79

our

-1.71

we

-1.69

 have

-1.65

 other

-1.61

 create

-1.59

ACCIÓN

-1.56

if

-1.53

？！

-1.52

POSITIVE LOGITS

</strong>

1.73

 современного

1.48

 quizás

1.41

 Perſ

1.41

referred

1.41

 ſont

1.36

 soooo

1.35

艶

1.35

 пять

1.34

 specif

1.34

Activations Density 0.057%