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

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

Neuron 3253 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.9227. Min Act: -0.0000

Data Index: 61596 (C4 (Web Text))

Max Activating Token Index: 550

Text #1

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.9106. Min Act: -0.0000

Data Index: 826024 (C4 (Web Text))

Max Activating Token Index: 873

Text #2

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.8939. Min Act: -0.0000

Data Index: 921754 (C4 (Web Text))

Max Activating Token Index: 436

Text #3

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.8651. Min Act: -0.0000

Data Index: 1087811 (C4 (Web Text))

Max Activating Token Index: 582

Text #4

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.8677. Min Act: -0.0000

Data Index: 926129 (C4 (Web Text))

Max Activating Token Index: 756

Text #5

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.8340. Min Act: -0.0000

Data Index: 263060 (C4 (Web Text))

Max Activating Token Index: 517

Text #6

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.7316. Min Act: -0.0000

Data Index: 500075 (C4 (Web Text))

Max Activating Token Index: 811

Text #7

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.8318. Min Act: -0.0000

Data Index: 1404259 (Python Code)

Max Activating Token Index: 468

Text #8

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6508. Min Act: -0.0000

Data Index: 286604 (C4 (Web Text))

Max Activating Token Index: 225

Text #9

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.7343. Min Act: -0.0000

Data Index: 1186360 (C4 (Web Text))

Max Activating Token Index: 268

Text #10

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.7346. Min Act: -0.0000

Data Index: 1526478 (Python Code)

Max Activating Token Index: 439

Text #11

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.7067. Min Act: -0.0000

Data Index: 138752 (C4 (Web Text))

Max Activating Token Index: 898

Text #12

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.7491. Min Act: -0.0000

Data Index: 969014 (C4 (Web Text))

Max Activating Token Index: 595

Text #13

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6985. Min Act: -0.0000

Data Index: 1514057 (Python Code)

Max Activating Token Index: 714

Text #14

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6788. Min Act: -0.0000

Data Index: 834269 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6562. Min Act: -0.0000

Data Index: 1565022 (Python Code)

Max Activating Token Index: 79

Text #16

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6179. Min Act: -0.0000

Data Index: 1414241 (Python Code)

Max Activating Token Index: 785

Text #17

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6595. Min Act: -0.0000

Data Index: 681069 (C4 (Web Text))

Max Activating Token Index: 590

Text #18

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6465. Min Act: -0.0000

Data Index: 565624 (C4 (Web Text))

Max Activating Token Index: 850

Text #19

Max Range: 0.9227. Min Range: -0.9227

Max Act: 0.6595. Min Act: -0.0000

Data Index: 1167687 (C4 (Web Text))

Max Activating Token Index: 573