INDEX

Explanations

unlike any other

The neuron detects comparative‐uniqueness constructions—phrases like “unlike any other,” “like no one,” or “nothing else” that emphasize something’s uniqueness.

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

</em>

-1.53

</strong>

-1.44

Such

-1.42

 就是

-1.35

oteur

-1.32

有时候

-1.30

<td>

-1.27

 冷凍

-1.25

)]);

-1.23

ropriate

-1.19

POSITIVE LOGITS

you

1.46

we

1.45

it

1.44

 also

1.39

 człowiek

1.26

 Campionato

1.24

 diğer

1.23

 that

1.23

 other

1.22

jednoc

1.20

Activations Density 0.011%