INDEX

Explanations

printing errors and alerts

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

 vecka

-0.97

 Conselho

-0.88

 prestazioni

-0.87

 jarang

-0.85

uje

-0.84

 PUSTAKA

-0.84

 suche

-0.84

ється

-0.83

 Perubahan

-0.83

legte

-0.82

POSITIVE LOGITS

1.20

you

1.17

ortiment

1.13

1.12

 plomo

1.11

 décrit

1.02

InjectAttribute

1.02

 nuevamente

1.00

něž

1.00

<>("

0.98

Activations Density 0.002%