INDEX

Explanations

best of both worlds

The neuron fires on occurrences of the phrase “both worlds” (as in “the best of both worlds”).

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

鯔

-1.41

ؓ

-1.39

 svar

-1.38

»;

-1.38

 labdar

-1.31

 COLOR

-1.30

</th>

-1.27

🤟

-1.27

 tärke

-1.22

 Interne

-1.22

POSITIVE LOGITS

۳

1.48

 не

1.46

 empe

1.42

都想

1.39

都不知道

1.38

 เป็น

1.36

first

1.35

 isso

1.34

…

1.33

mathbf

1.31

Activations Density 0.017%