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

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

Neuron 2026 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.7538. Min Act: -0.0000

Data Index: 891962 (C4 (Web Text))

Max Activating Token Index: 139

Text #1

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.4761. Min Act: -0.0000

Data Index: 926842 (C4 (Web Text))

Max Activating Token Index: 165

Text #2

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.5643. Min Act: -0.0000

Data Index: 825714 (C4 (Web Text))

Max Activating Token Index: 243

Text #3

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.6406. Min Act: -0.0000

Data Index: 840944 (C4 (Web Text))

Max Activating Token Index: 954

Text #4

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.3412. Min Act: -0.0000

Data Index: 1399358 (Python Code)

Max Activating Token Index: 243

Text #5

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1528. Min Act: -0.0000

Data Index: 52854 (C4 (Web Text))

Max Activating Token Index: 228

Text #6

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.3178. Min Act: -0.0000

Data Index: 1067605 (C4 (Web Text))

Max Activating Token Index: 412

Text #7

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.2742. Min Act: -0.0000

Data Index: 749205 (C4 (Web Text))

Max Activating Token Index: 417

Text #8

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.2125. Min Act: -0.0000

Data Index: 739715 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.3251. Min Act: -0.0000

Data Index: 1603449 (Python Code)

Max Activating Token Index: 654

Text #10

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1030. Min Act: -0.0000

Data Index: 192005 (C4 (Web Text))

Max Activating Token Index: 224

Text #11

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1052. Min Act: -0.0000

Data Index: 1163986 (C4 (Web Text))

Max Activating Token Index: 256

Text #12

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1929. Min Act: -0.0000

Data Index: 1007460 (C4 (Web Text))

Max Activating Token Index: 63

Text #13

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.2200. Min Act: -0.0000

Data Index: 210038 (C4 (Web Text))

Max Activating Token Index: 617

Text #14

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.0667. Min Act: -0.0000

Data Index: 1479266 (Python Code)

Max Activating Token Index: 390

Text #15

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.0366. Min Act: -0.0000

Data Index: 1379508 (Python Code)

Max Activating Token Index: 551

Text #16

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1448. Min Act: -0.0000

Data Index: 562475 (C4 (Web Text))

Max Activating Token Index: 566

Text #17

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.1282. Min Act: -0.0000

Data Index: 1611066 (Python Code)

Max Activating Token Index: 1007

Text #18

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.0163. Min Act: -0.0000

Data Index: 1595773 (Python Code)

Max Activating Token Index: 254

Text #19

Max Range: 2.7538. Min Range: -2.7538

Max Act: 2.0054. Min Act: -0.0000

Data Index: 102980 (C4 (Web Text))

Max Activating Token Index: 894