Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3670 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.9751. Min Act: -0.0000

Data Index: 1451465 (Python Code)

Max Activating Token Index: 736

Text #1

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.8167. Min Act: -0.0000

Data Index: 1152858 (C4 (Web Text))

Max Activating Token Index: 947

Text #2

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.8913. Min Act: -0.0000

Data Index: 1540667 (Python Code)

Max Activating Token Index: 428

Text #3

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.8898. Min Act: -0.0000

Data Index: 804819 (C4 (Web Text))

Max Activating Token Index: 462

Text #4

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5572. Min Act: -0.0000

Data Index: 1017769 (C4 (Web Text))

Max Activating Token Index: 539

Text #5

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.6213. Min Act: -0.0000

Data Index: 193561 (C4 (Web Text))

Max Activating Token Index: 1007

Text #6

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.6022. Min Act: -0.0000

Data Index: 408639 (C4 (Web Text))

Max Activating Token Index: 1004

Text #7

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5666. Min Act: -0.0000

Data Index: 999885 (C4 (Web Text))

Max Activating Token Index: 659

Text #8

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.3753. Min Act: -0.0000

Data Index: 754546 (C4 (Web Text))

Max Activating Token Index: 391

Text #9

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5118. Min Act: -0.0000

Data Index: 147351 (C4 (Web Text))

Max Activating Token Index: 437

Text #10

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4972. Min Act: -0.0000

Data Index: 275237 (C4 (Web Text))

Max Activating Token Index: 461

Text #11

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5300. Min Act: -0.0000

Data Index: 1644398 (Python Code)

Max Activating Token Index: 275

Text #12

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5117. Min Act: -0.0000

Data Index: 209715 (C4 (Web Text))

Max Activating Token Index: 729

Text #13

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4946. Min Act: -0.0000

Data Index: 424879 (C4 (Web Text))

Max Activating Token Index: 260

Text #14

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4987. Min Act: -0.0000

Data Index: 631340 (C4 (Web Text))

Max Activating Token Index: 201

Text #15

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.5000. Min Act: -0.0000

Data Index: 419269 (C4 (Web Text))

Max Activating Token Index: 616

Text #16

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4993. Min Act: -0.0000

Data Index: 1359094 (C4 (Web Text))

Max Activating Token Index: 850

Text #17

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.3501. Min Act: -0.0000

Data Index: 816269 (C4 (Web Text))

Max Activating Token Index: 699

Text #18

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4328. Min Act: -0.0000

Data Index: 1609365 (Python Code)

Max Activating Token Index: 717

Text #19

Max Range: 2.9751. Min Range: -2.9751

Max Act: 2.4644. Min Act: -0.0000

Data Index: 409315 (C4 (Web Text))

Max Activating Token Index: 872