INDEX

Explanations

forth

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 forth

-0.95

forth

-0.94

TagMode

-0.94

paravant

-0.81

moiselle

-0.80

 myſelf

-0.79

 виправивши

-0.74

 disambiguazione

-0.73

Rohy

-0.73

 Monfieur

-0.72

POSITIVE LOGITS

ngrx

0.52

of

0.48

Dep

0.43

how

0.41

dep

0.41

our

0.37

 more

0.37

dep

0.36

com

0.35

 general

0.35

Activations Density 0.024%