INDEX

Explanations

mathematical symbols and notations

This neuron never activates on any of the tokens shown—it appears to be effectively inactive and not detecting any pattern.

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

ίη

-1.24

ﺅ

-1.22

rir

-1.19

}}.

-1.15

 colorir

-1.15

Dieser

-1.13

زیین

-1.13

他们

-1.13

𐍆

-1.12

 Anſ

-1.11

POSITIVE LOGITS

 having

1.15

for

1.10

 with

1.10

 Since

1.05

any

1.05

🩹

1.03

 because

1.02

Since

1.00

 vulgares

0.99

Ἔ

0.99

Activations Density 0.004%