INDEX

Explanations

title: or quotation marks

This neuron detects opening parenthesis tokens, i.e. “(”, marking parenthetical material or citations.

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

</h4>

-2.13

 Ensure

-1.88

После

-1.82

 Undoubtedly

-1.79

 oleju

-1.77

Semua

-1.76

Что

-1.74

</strong>

-1.74

 Presumably

-1.73

 всіх

-1.71

POSITIVE LOGITS

the

2.05

 they

1.95

 when

1.80

锨

1.68

we

1.67

有任何

1.66

\.

1.63

谖

1.63

}*/

1.57

."

1.55

Activations Density 0.014%