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

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

Neuron 1727 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: 6.7198. Min Range: -6.7198

Max Act: 6.7198. Min Act: -0.0000

Data Index: 1586276 (Python Code)

Max Activating Token Index: 168

Text #1

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.7356. Min Act: -0.0000

Data Index: 1091594 (C4 (Web Text))

Max Activating Token Index: 1017

Text #2

Max Range: 6.7198. Min Range: -6.7198

Max Act: 5.5609. Min Act: -0.0000

Data Index: 568191 (C4 (Web Text))

Max Activating Token Index: 368

Text #3

Max Range: 6.7198. Min Range: -6.7198

Max Act: 5.1166. Min Act: -0.0000

Data Index: 1073688 (C4 (Web Text))

Max Activating Token Index: 462

Text #4

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.6025. Min Act: -0.0000

Data Index: 339819 (C4 (Web Text))

Max Activating Token Index: 368

Text #5

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.8578. Min Act: -0.0000

Data Index: 400552 (C4 (Web Text))

Max Activating Token Index: 417

Text #6

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.8308. Min Act: -0.0000

Data Index: 685563 (C4 (Web Text))

Max Activating Token Index: 955

Text #7

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.3038. Min Act: -0.0000

Data Index: 31182 (C4 (Web Text))

Max Activating Token Index: 208

Text #8

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.5321. Min Act: -0.0000

Data Index: 1444700 (Python Code)

Max Activating Token Index: 933

Text #9

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.4608. Min Act: -0.0000

Data Index: 1457964 (Python Code)

Max Activating Token Index: 255

Text #10

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.7978. Min Act: -0.0000

Data Index: 1218738 (C4 (Web Text))

Max Activating Token Index: 1008

Text #11

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.2900. Min Act: -0.0000

Data Index: 1052257 (C4 (Web Text))

Max Activating Token Index: 694

Text #12

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.4232. Min Act: -0.0000

Data Index: 549225 (C4 (Web Text))

Max Activating Token Index: 842

Text #13

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.4549. Min Act: -0.0000

Data Index: 749325 (C4 (Web Text))

Max Activating Token Index: 707

Text #14

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.0621. Min Act: -0.0000

Data Index: 999943 (C4 (Web Text))

Max Activating Token Index: 847

Text #15

Max Range: 6.7198. Min Range: -6.7198

Max Act: 3.3917. Min Act: -0.0000

Data Index: 1014154 (C4 (Web Text))

Max Activating Token Index: 947

Text #16

Max Range: 6.7198. Min Range: -6.7198

Max Act: 3.5279. Min Act: -0.0000

Data Index: 900158 (C4 (Web Text))

Max Activating Token Index: 391

Text #17

Max Range: 6.7198. Min Range: -6.7198

Max Act: 3.5910. Min Act: -0.0000

Data Index: 1509589 (Python Code)

Max Activating Token Index: 335

Text #18

Max Range: 6.7198. Min Range: -6.7198

Max Act: 4.2888. Min Act: -0.0000

Data Index: 194357 (C4 (Web Text))

Max Activating Token Index: 726

Text #19

Max Range: 6.7198. Min Range: -6.7198

Max Act: 3.2277. Min Act: -0.0000

Data Index: 688403 (C4 (Web Text))

Max Activating Token Index: 372