INDEX

Explanations

Power prefix

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

 trecho

-0.83

 começaram

-0.82

cuisine

-0.82

 Nobles

-0.76

Cuisine

-0.75

turismo

-0.75

ratios

-0.75

recognition

-0.73

 Cuisine

-0.71

 devez

-0.70

POSITIVE LOGITS

 Power

1.58

Power

1.54

 powerless

1.41

 power

1.38

power

1.23

POWER

1.13

Powered

1.13

 POWER

1.09

Mockito

1.05

powered

0.90

Activations Density 0.029%