INDEX

Explanations

list types and angle brackets

The neuron activates on tokens containing numeric digits (e.g. numeric literals or identifiers with numbers).

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

 When

-1.32

—

-1.23

These

-1.22

If

-1.21

-1.19

//}

-1.15

 During

-1.15

 Using

-1.14

 should

-1.13

There

-1.13

POSITIVE LOGITS

もあるので

1.31

 plazos

1.27

 détect

1.23

もあって

1.17

 laisser

1.16

 maî

1.16

だけでなく

1.14

[]

1.13

もあり

1.13

 miſ

1.13

Activations Density 0.006%