INDEX

Explanations

as possible

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

 last

-1.05

最后一个

-1.03

yaç

-0.90

｝

-0.90

 vena

-0.88

ANTS

-0.88

 таким

-0.86

 prenda

-0.85

RTHOOK

-0.85

 ロゴ

-0.85

POSITIVE LOGITS

 possible

2.91

possible

1.98

as

1.95

 posible

1.94

Possible

1.91

 Possible

1.91

 mogelijk

1.91

 possível

1.76

 posibles

1.64

POSSIBLE

1.61

Activations Density 0.012%