INDEX

Explanations

The neuron detects reference or citation markers (e.g. superscript arrows and numbers) in academic‐style bibliographic entries.

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

ər

0.71

𝐕

0.71

 其他

0.67

0.66

𝙫

0.65

0.64

 nữ

0.64

囪

0.64

 احد

0.64

 sırada

0.64

POSITIVE LOGITS

 لهذا

0.61

 γίνει

0.58

 ويكي

0.57

 لهذه

0.56

繊維

0.55

 diperoleh

0.55

 निश्च

0.55

>{</

0.55

 legislação

0.54

 authors

0.54

Activations Density 0.002%