INDEX

Explanations

periods of time

The neuron detects mentions of “time” or other explicit temporal references.

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

ྂ

-1.45

するか

-1.41

 first

-1.39

 vorhandenen

-1.38

 from

-1.36

-1.34

thanx

-1.33

歿

-1.33

llac

-1.30

 entsprechenden

-1.29

POSITIVE LOGITS

 между

1.46

in

1.38

 Đình

1.34



1.32

 capitale

1.30

lántico

1.27

 ギャラリー

1.26

佐賀

1.25

 energy

1.23

estión

1.23

Activations Density 0.027%