INDEX

Explanations

is the [noun]

The neuron consistently detects the definite article “the.”

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

-2.13

 عندما

-1.80

}{

-1.70

-1.67

ראלי

-1.62

had

-1.62

 günlük

-1.60

 όταν

-1.60

 become

-1.57

 llamadas

-1.57

POSITIVE LOGITS



1.96

 beve

1.88

鲼

1.86

?’

1.81

 nime

1.80

currently

1.78

 Startseite

1.77

 STILL

1.76

 kere

1.75

大家好

1.74

Activations Density 0.019%