INDEX

Explanations

opening quoted titles

The neuron detects opening quotation marks (i.e. the beginning of quoted titles or phrases).

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

可能會

-1.63

igrette

-1.60

 dificultades

-1.52

 productie

-1.50

 ""),

-1.45

]);

-1.41

 fantástico

-1.41

ſſen

-1.41

ditionally

-1.41

 می‌شود

-1.41

POSITIVE LOGITS

for

2.08

and

1.94

ಞ

1.81

st

1.70

 Also

1.66

in

1.55

ie

1.51

net

1.48

}\

1.47

ఞ

1.44

Activations Density 0.001%