INDEX

Explanations

of course

This neuron detects instances of the phrase “Of course.”

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

and

-2.77

\},

-1.84

}^{

-1.84

でも

-1.81

niająca

-1.80

}$.

-1.79

ER

-1.72

to

-1.71

プター

-1.66

 просто

-1.65

POSITIVE LOGITS

 maga

1.67

ing

1.61

 they

1.59

橚

1.57

 Islas

1.55

にほんブログ村

1.53

atro

1.51

ᄋ

1.48

𝑁

1.48

𝒌

1.48

Activations Density 0.006%