INDEX

Explanations

back into / on

This neuron activates on uses of the word “back” when it appears in phrasal constructions (e.g. “back into,” “back on,” “back on track”).

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

 also

-1.60

 bereits

-1.50

 there

-1.50

 également

-1.48

and

-1.35

 уже

-1.34

 listos

-1.28

 zároveň

-1.27

 nooit

-1.25

it

-1.23

POSITIVE LOGITS

 tuvimos

1.57

 teníamos

1.43

}`;

1.29

 começaram

1.28

who

1.28

をご紹介します

1.27

↵

1.27

1.26

 '".$

1.23

})

1.22

Activations Density 0.071%