INDEX

Explanations

instead of followed by verb-ing

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

-2.27

↵↵

-2.17

-1.99

-1.98

所有

-1.97

-1.96

);

-1.95

给人

-1.83

 Уже

-1.82

];

-1.81

POSITIVE LOGITS

 have

2.19

 désol

1.95

CCIÓN

1.88

 They

1.85

炱

1.84

 которой

1.83

 İyi

1.81

 særlig

1.80

NESSEE

1.78

сшта

1.77

Activations Density 0.010%