INDEX

Explanations

Op Capability / Pain Relief

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

our

-1.97

 lesquelles

-1.81

bawa

-1.62

-1.52

și

-1.52

 expédié

-1.51

of

-1.48

𝐒

-1.46

沨

-1.45

fotograf

-1.40

POSITIVE LOGITS

↵↵

1.95

in

1.92

).

1.77

 оригі

1.70

 there

1.60

has

1.51

癭

1.50

are

1.49

pography

1.38

 still

1.37

Activations Density 0.003%