INDEX

Explanations

.as(

This neuron fires on code-specific keywords, particularly the tokens “from” and “as.”

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

 FormBuilder

-1.42

 сентября

-1.36

...

-1.34

 dubbed

-1.33

 broek

-1.33

 then

-1.33

 gră

-1.33

が使

-1.33

end

-1.30

</h1>

-1.30

POSITIVE LOGITS

of

2.73

you

1.77

While

1.62

 Owing

1.50

These

1.49

Though

1.45

красивые

1.45

warnai

1.44

 These

1.40

 Lately

1.38

Activations Density 0.000%