INDEX

Explanations

coercive action, questioning, or government influence

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

इटम

-3.41

祧

-2.75

</h3>

-2.73

\"

-2.56

廔

-2.48

ſſen

-2.42

 }));

-2.38

 ",";

-2.36

)();

-2.28

 tragically

-2.27

POSITIVE LOGITS

</b>

3.13

↵

2.92

2.48

2.25

味噌汁

2.05

↵↵↵

2.02

 forsaken

1.98

1.97

 шанс

1.96

1.93

Activations Density 0.006%