INDEX

Explanations

what's more, worse, better

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

 lainnya

-1.05

)}</

-1.02

何を

-1.02

obviously

-1.00

``.

-0.98

 áng

-0.97

 berbasis

-0.97

]="

-0.96

}>;

-0.95

 sám

-0.95

POSITIVE LOGITS

 حتی

1.76

 even

1.76

 sogar

1.72

 dokonce

1.54

 даже

1.36

 jopa

1.36

 Worse

1.33

 nawet

1.33

さらに

1.26

 worse

1.22

Activations Density 0.018%