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

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

Neuron 2715 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: 4.1354. Min Range: -4.1354

Max Act: 3.4280. Min Act: -0.0000

Data Index: 1497029 (Python Code)

Max Activating Token Index: 179

Text #1

Max Range: 4.1354. Min Range: -4.1354

Max Act: 4.1354. Min Act: -0.0000

Data Index: 225929 (C4 (Web Text))

Max Activating Token Index: 837

Text #2

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.6198. Min Act: -0.0000

Data Index: 858610 (C4 (Web Text))

Max Activating Token Index: 62

Text #3

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.4274. Min Act: -0.0000

Data Index: 1496502 (Python Code)

Max Activating Token Index: 326

Text #4

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.5799. Min Act: -0.0000

Data Index: 490394 (C4 (Web Text))

Max Activating Token Index: 945

Text #5

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.7229. Min Act: -0.0000

Data Index: 1089896 (C4 (Web Text))

Max Activating Token Index: 614

Text #6

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.6611. Min Act: -0.0000

Data Index: 557565 (C4 (Web Text))

Max Activating Token Index: 595

Text #7

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.4236. Min Act: -0.0000

Data Index: 1335893 (C4 (Web Text))

Max Activating Token Index: 4

Text #8

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.3878. Min Act: -0.0000

Data Index: 1629378 (Python Code)

Max Activating Token Index: 554

Text #9

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.1755. Min Act: -0.0000

Data Index: 481104 (C4 (Web Text))

Max Activating Token Index: 35

Text #10

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.6867. Min Act: -0.0000

Data Index: 1105193 (C4 (Web Text))

Max Activating Token Index: 713

Text #11

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.2730. Min Act: -0.0000

Data Index: 345594 (C4 (Web Text))

Max Activating Token Index: 813

Text #12

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.3651. Min Act: -0.0000

Data Index: 1503675 (Python Code)

Max Activating Token Index: 997

Text #13

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.2644. Min Act: -0.0000

Data Index: 920722 (C4 (Web Text))

Max Activating Token Index: 1007

Text #14

Max Range: 4.1354. Min Range: -4.1354

Max Act: 2.9312. Min Act: -0.0000

Data Index: 175941 (C4 (Web Text))

Max Activating Token Index: 314

Text #15

Max Range: 4.1354. Min Range: -4.1354

Max Act: 2.8235. Min Act: -0.0000

Data Index: 1276000 (C4 (Web Text))

Max Activating Token Index: 840

Text #16

Max Range: 4.1354. Min Range: -4.1354

Max Act: 2.8129. Min Act: -0.0000

Data Index: 158973 (C4 (Web Text))

Max Activating Token Index: 929

Text #17

Max Range: 4.1354. Min Range: -4.1354

Max Act: 3.0932. Min Act: -0.0000

Data Index: 352176 (C4 (Web Text))

Max Activating Token Index: 888

Text #18

Max Range: 4.1354. Min Range: -4.1354

Max Act: 2.9179. Min Act: -0.0000

Data Index: 1086159 (C4 (Web Text))

Max Activating Token Index: 769

Text #19

Max Range: 4.1354. Min Range: -4.1354

Max Act: 2.7520. Min Act: -0.0000

Data Index: 815268 (C4 (Web Text))

Max Activating Token Index: 175