Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2535 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5675. Min Act: -0.0001

Data Index: 1372165 (Python Code)

Max Activating Token Index: 940

Text #1

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5392. Min Act: -0.0001

Data Index: 63492 (C4 (Web Text))

Max Activating Token Index: 784

Text #2

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5045. Min Act: -0.0001

Data Index: 942086 (C4 (Web Text))

Max Activating Token Index: 369

Text #3

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5133. Min Act: -0.0001

Data Index: 381553 (C4 (Web Text))

Max Activating Token Index: 511

Text #4

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5099. Min Act: -0.0001

Data Index: 171177 (C4 (Web Text))

Max Activating Token Index: 946

Text #5

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5143. Min Act: -0.0001

Data Index: 1214761 (C4 (Web Text))

Max Activating Token Index: 622

Text #6

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.5241. Min Act: -0.0001

Data Index: 475519 (C4 (Web Text))

Max Activating Token Index: 175

Text #7

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4842. Min Act: -0.0001

Data Index: 212119 (C4 (Web Text))

Max Activating Token Index: 116

Text #8

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4622. Min Act: -0.0001

Data Index: 180354 (C4 (Web Text))

Max Activating Token Index: 273

Text #9

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4472. Min Act: -0.0001

Data Index: 1028246 (C4 (Web Text))

Max Activating Token Index: 173

Text #10

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4730. Min Act: -0.0001

Data Index: 941493 (C4 (Web Text))

Max Activating Token Index: 18

Text #11

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4579. Min Act: -0.0001

Data Index: 417829 (C4 (Web Text))

Max Activating Token Index: 153

Text #12

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4707. Min Act: -0.0001

Data Index: 1364026 (Python Code)

Max Activating Token Index: 6

Text #13

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4703. Min Act: -0.0001

Data Index: 260546 (C4 (Web Text))

Max Activating Token Index: 750

Text #14

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4662. Min Act: -0.0001

Data Index: 789624 (C4 (Web Text))

Max Activating Token Index: 748

Text #15

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4619. Min Act: -0.0001

Data Index: 828339 (C4 (Web Text))

Max Activating Token Index: 973

Text #16

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4740. Min Act: -0.0001

Data Index: 132979 (C4 (Web Text))

Max Activating Token Index: 691

Text #17

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4604. Min Act: -0.0001

Data Index: 359566 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4533. Min Act: -0.0001

Data Index: 1487715 (Python Code)

Max Activating Token Index: 956

Text #19

Max Range: 0.5675. Min Range: -0.5675

Max Act: 0.4643. Min Act: -0.0001

Data Index: 949963 (C4 (Web Text))

Max Activating Token Index: 18