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

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

Neuron 5510 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: 4.6372. Min Range: -4.6372

Max Act: 4.3805. Min Act: -0.0000

Data Index: 949746 (C4 (Web Text))

Max Activating Token Index: 215

Text #1

Max Range: 4.6372. Min Range: -4.6372

Max Act: 4.6372. Min Act: -0.0000

Data Index: 82944 (C4 (Web Text))

Max Activating Token Index: 937

Text #2

Max Range: 4.6372. Min Range: -4.6372

Max Act: 3.8352. Min Act: -0.0000

Data Index: 156502 (C4 (Web Text))

Max Activating Token Index: 395

Text #3

Max Range: 4.6372. Min Range: -4.6372

Max Act: 3.2453. Min Act: -0.0000

Data Index: 83390 (C4 (Web Text))

Max Activating Token Index: 488

Text #4

Max Range: 4.6372. Min Range: -4.6372

Max Act: 3.0125. Min Act: -0.0000

Data Index: 606540 (C4 (Web Text))

Max Activating Token Index: 663

Text #5

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.6121. Min Act: -0.0000

Data Index: 1094764 (C4 (Web Text))

Max Activating Token Index: 752

Text #6

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.5651. Min Act: -0.0000

Data Index: 223742 (C4 (Web Text))

Max Activating Token Index: 774

Text #7

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.3450. Min Act: -0.0000

Data Index: 753850 (C4 (Web Text))

Max Activating Token Index: 983

Text #8

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.4767. Min Act: -0.0000

Data Index: 526310 (C4 (Web Text))

Max Activating Token Index: 468

Text #9

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.3192. Min Act: -0.0000

Data Index: 409139 (C4 (Web Text))

Max Activating Token Index: 193

Text #10

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.3053. Min Act: -0.0000

Data Index: 1105821 (C4 (Web Text))

Max Activating Token Index: 664

Text #11

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.2911. Min Act: -0.0000

Data Index: 206886 (C4 (Web Text))

Max Activating Token Index: 756

Text #12

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.2692. Min Act: -0.0000

Data Index: 1622035 (Python Code)

Max Activating Token Index: 267

Text #13

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.2677. Min Act: 0.0000

Data Index: 1318909 (C4 (Web Text))

Max Activating Token Index: 83

Text #14

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.1445. Min Act: -0.0000

Data Index: 888961 (C4 (Web Text))

Max Activating Token Index: 830

Text #15

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.0689. Min Act: -0.0000

Data Index: 1048356 (C4 (Web Text))

Max Activating Token Index: 918

Text #16

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.1118. Min Act: -0.0000

Data Index: 701624 (C4 (Web Text))

Max Activating Token Index: 654

Text #17

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.1248. Min Act: -0.0000

Data Index: 1020838 (C4 (Web Text))

Max Activating Token Index: 510

Text #18

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.1247. Min Act: -0.0000

Data Index: 1353513 (C4 (Web Text))

Max Activating Token Index: 661

Text #19

Max Range: 4.6372. Min Range: -4.6372

Max Act: 2.0019. Min Act: -0.0000

Data Index: 222764 (C4 (Web Text))

Max Activating Token Index: 752