INDEX

Explanations

other, another, later, again

This neuron detects academic citation markers and references (e.g., “et al.”, author names, years, bracketed reference numbers).

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

 zelfs

-0.95

TypeArgs

-0.94

 entspricht

-0.90

 nenhuma

-0.89

Slik

-0.86

纐

-0.85

 zarówno

-0.84

 endlich

-0.83

が含ま

-0.83

 पूरी

-0.82

POSITIVE LOGITS

 later

1.52

 another

1.43

 autre

1.30

另外

1.22

 другой

1.17

 other

1.13

 also

1.12

later

1.12

 másik

1.08

 again

1.05

Activations Density 0.027%