Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3527 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.7703. Min Act: -0.0000

Data Index: 1513048 (Python Code)

Max Activating Token Index: 127

Text #1

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.7476. Min Act: -0.0000

Data Index: 1603336 (Python Code)

Max Activating Token Index: 197

Text #2

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.7397. Min Act: -0.0000

Data Index: 1492145 (Python Code)

Max Activating Token Index: 358

Text #3

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.7228. Min Act: -0.0000

Data Index: 774038 (C4 (Web Text))

Max Activating Token Index: 926

Text #4

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6607. Min Act: -0.0000

Data Index: 1564797 (Python Code)

Max Activating Token Index: 125

Text #5

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6552. Min Act: -0.0000

Data Index: 426968 (C4 (Web Text))

Max Activating Token Index: 748

Text #6

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6753. Min Act: -0.0000

Data Index: 327860 (C4 (Web Text))

Max Activating Token Index: 951

Text #7

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6725. Min Act: -0.0000

Data Index: 622604 (C4 (Web Text))

Max Activating Token Index: 423

Text #8

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6223. Min Act: -0.0000

Data Index: 1289185 (C4 (Web Text))

Max Activating Token Index: 680

Text #9

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6790. Min Act: -0.0000

Data Index: 465939 (C4 (Web Text))

Max Activating Token Index: 446

Text #10

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.5995. Min Act: -0.0000

Data Index: 683306 (C4 (Web Text))

Max Activating Token Index: 759

Text #11

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6296. Min Act: -0.0000

Data Index: 1491087 (Python Code)

Max Activating Token Index: 383

Text #12

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6384. Min Act: -0.0000

Data Index: 1546457 (Python Code)

Max Activating Token Index: 156

Text #13

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6215. Min Act: -0.0000

Data Index: 322852 (C4 (Web Text))

Max Activating Token Index: 848

Text #14

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6146. Min Act: -0.0000

Data Index: 1422387 (Python Code)

Max Activating Token Index: 35

Text #15

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6357. Min Act: -0.0000

Data Index: 1015655 (C4 (Web Text))

Max Activating Token Index: 299

Text #16

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6312. Min Act: -0.0000

Data Index: 1630917 (Python Code)

Max Activating Token Index: 254

Text #17

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.5853. Min Act: -0.0000

Data Index: 97736 (C4 (Web Text))

Max Activating Token Index: 763

Text #18

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6237. Min Act: -0.0000

Data Index: 314639 (C4 (Web Text))

Max Activating Token Index: 229

Text #19

Max Range: 0.7703. Min Range: -0.7703

Max Act: 0.6428. Min Act: -0.0000

Data Index: 1458051 (Python Code)

Max Activating Token Index: 116