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

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

Neuron 4084 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: 2.8925. Min Range: -2.8925

Max Act: 2.8925. Min Act: -0.0000

Data Index: 1473910 (Python Code)

Max Activating Token Index: 894

Text #1

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.6735. Min Act: -0.0000

Data Index: 1505699 (Python Code)

Max Activating Token Index: 863

Text #2

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.5378. Min Act: -0.0000

Data Index: 246778 (C4 (Web Text))

Max Activating Token Index: 148

Text #3

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.7448. Min Act: -0.0000

Data Index: 1124536 (C4 (Web Text))

Max Activating Token Index: 620

Text #4

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.5317. Min Act: -0.0000

Data Index: 1647620 (Python Code)

Max Activating Token Index: 641

Text #5

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.4019. Min Act: -0.0000

Data Index: 1563062 (Python Code)

Max Activating Token Index: 194

Text #6

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.3815. Min Act: -0.0000

Data Index: 1431630 (Python Code)

Max Activating Token Index: 668

Text #7

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.3727. Min Act: -0.0000

Data Index: 82578 (C4 (Web Text))

Max Activating Token Index: 31

Text #8

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.2413. Min Act: -0.0000

Data Index: 432971 (C4 (Web Text))

Max Activating Token Index: 296

Text #9

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.3626. Min Act: -0.0000

Data Index: 285850 (C4 (Web Text))

Max Activating Token Index: 514

Text #10

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.2764. Min Act: -0.0000

Data Index: 1500891 (Python Code)

Max Activating Token Index: 334

Text #11

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.0897. Min Act: -0.0000

Data Index: 103366 (C4 (Web Text))

Max Activating Token Index: 358

Text #12

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.2030. Min Act: -0.0000

Data Index: 1600936 (Python Code)

Max Activating Token Index: 670

Text #13

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.3222. Min Act: -0.0000

Data Index: 1445399 (Python Code)

Max Activating Token Index: 286

Text #14

Max Range: 2.8925. Min Range: -2.8925

Max Act: 1.9765. Min Act: -0.0000

Data Index: 292048 (C4 (Web Text))

Max Activating Token Index: 796

Text #15

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.1239. Min Act: -0.0000

Data Index: 182583 (C4 (Web Text))

Max Activating Token Index: 956

Text #16

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.0916. Min Act: -0.0000

Data Index: 1450310 (Python Code)

Max Activating Token Index: 417

Text #17

Max Range: 2.8925. Min Range: -2.8925

Max Act: 1.9495. Min Act: -0.0000

Data Index: 899174 (C4 (Web Text))

Max Activating Token Index: 746

Text #18

Max Range: 2.8925. Min Range: -2.8925

Max Act: 2.0688. Min Act: -0.0000

Data Index: 248383 (C4 (Web Text))

Max Activating Token Index: 277

Text #19

Max Range: 2.8925. Min Range: -2.8925

Max Act: 1.9263. Min Act: -0.0000

Data Index: 921384 (C4 (Web Text))

Max Activating Token Index: 187