INDEX

Explanations

positive sentiment and appreciation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abort

0.77

 बवाल

0.75

 заседа

0.73

 disastrous

0.72

 ужа

0.71

 berbahaya

0.71

 eyelids

0.71

 sorular

0.70

 நிர்

0.69

 Alarm

0.69

POSITIVE LOGITS

 appreciation

1.97

 recognition

1.88

 Recognition

1.79

 Appreciation

1.78

 praise

1.78

Recognition

1.78

 acknowledgement

1.76

 acknowledgment

1.76

 gratitude

1.72

 praises

1.69

Activations Density 0.760%