INDEX

Explanations

foreign phrases and words

The neuron fires on two‐letter syllables made up of a consonant plus “e” or “o” (e.g. “Le”, “de”, “do”).

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

</u>

-1.80

</strong>

-1.79

to

-1.55

</h3>

-1.55

“

-1.49

</i>

-1.48

and

-1.45

 REGUL

-1.42

such

-1.38

 repre

-1.36

POSITIVE LOGITS

ệu

1.46

 Vordergrund

1.42

чего

1.39

ztály

1.38

 monasterio

1.38

 commento

1.37

ülés

1.37

 Ucrania

1.35

gucig

1.34

ARON

1.34

Activations Density 0.001%