INDEX

Explanations

may or may not / more or less

This neuron detects modal‐uncertainty constructions—specifically the “may … or …” pattern that expresses optionality.

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

of

-2.75

このように

-1.90

 members

-1.90

 sizable

-1.89

).

-1.83

 realizacji

-1.80

 sizeable

-1.80

ACIONAL

-1.80

可以直接

-1.75

 незавер

-1.73

POSITIVE LOGITS

⟣

1.98

 also

1.89

ons

1.88

1.80

*/

1.67

..,

1.66

禟

1.59

volle

1.57

 WHILE

1.54

chtige

1.52

Activations Density 0.015%