INDEX

Explanations

intensifying adverbs

The neuron lights up on adverbs and other qualifiers that add emphasis or modality (e.g. “endlessly,” “automatically,” “possible,” “easily,” “done,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 επίσης

0.98

이었

0.96

 середине

0.90

 апреле

0.86

⌓

0.86

+"\

0.84

 വരെ

0.82

 しかし

0.82

𝗜

0.82

 တော့

0.82

POSITIVE LOGITS

 andere

0.95

នូវ

0.83

the

0.80

arene

0.78

 other

0.77

เกี่ยวกับ

0.76

 anderen

0.75

0.74

その他の

0.73

Activations Density 0.922%