INDEX

Explanations

blessing

The main thing this neuron does is detect words and phrases expressing a blessing (e.g., “bless,” “blessing,” “blessed”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

</strong>

-2.14

⢪

-2.02

</h3>

-1.87

-1.80

獰

-1.75

calipsis

-1.73

឵

-1.73

筇

-1.71

 кухо

-1.70

ンバー

-1.69

POSITIVE LOGITS

贗

2.09

1.91

 speciale

1.84

kembangkan

1.78

新型

1.70

墊

1.67

谖

1.63

 stata

1.61

This

1.60

摃

1.60

Activations Density 0.007%