INDEX

Explanations

Roman Catholic and historical references

This neuron detects mentions of “Roman” (and closely related cultural or historical terms like Greco-Roman, Latin, Catholic Church, etc.).

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

偁

-1.71

:「

-1.65

𓇼

-1.56

 vývoj

-1.52

sibilité

-1.51

玏

-1.51

apatalk

-1.48

hysema

-1.46

惕

-1.46

已经

-1.45

POSITIVE LOGITS

of

1.57

 giacca

1.49

ar

1.48

わが

1.43

 detrimental

1.41

dziel

1.38

着一

1.35

 inconveniente

1.29

 toalha

1.29

</h5>

1.28

Activations Density 0.039%