Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 23 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0460. Min Act: -0.0003

Data Index: 351837 (C4 (Web Text))

Max Activating Token Index: 114

Text #1

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0463. Min Act: -0.0003

Data Index: 228978 (C4 (Web Text))

Max Activating Token Index: 154

Text #2

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0445. Min Act: -0.0003

Data Index: 25303 (C4 (Web Text))

Max Activating Token Index: 541

Text #3

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0440. Min Act: -0.0003

Data Index: 1169960 (C4 (Web Text))

Max Activating Token Index: 302

Text #4

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0451. Min Act: -0.0004

Data Index: 929512 (C4 (Web Text))

Max Activating Token Index: 118

Text #5

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0442. Min Act: -0.0003

Data Index: 636296 (C4 (Web Text))

Max Activating Token Index: 475

Text #6

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0430. Min Act: -0.0003

Data Index: 290372 (C4 (Web Text))

Max Activating Token Index: 147

Text #7

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0431. Min Act: -0.0003

Data Index: 960209 (C4 (Web Text))

Max Activating Token Index: 847

Text #8

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0438. Min Act: -0.0003

Data Index: 772105 (C4 (Web Text))

Max Activating Token Index: 849

Text #9

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0439. Min Act: -0.0003

Data Index: 839820 (C4 (Web Text))

Max Activating Token Index: 550

Text #10

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0431. Min Act: -0.0003

Data Index: 697950 (C4 (Web Text))

Max Activating Token Index: 803

Text #11

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0432. Min Act: -0.0003

Data Index: 187614 (C4 (Web Text))

Max Activating Token Index: 72

Text #12

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0436. Min Act: -0.0003

Data Index: 298873 (C4 (Web Text))

Max Activating Token Index: 335

Text #13

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0423. Min Act: -0.0003

Data Index: 866778 (C4 (Web Text))

Max Activating Token Index: 670

Text #14

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0431. Min Act: -0.0003

Data Index: 1167254 (C4 (Web Text))

Max Activating Token Index: 444

Text #15

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0418. Min Act: -0.0003

Data Index: 552146 (C4 (Web Text))

Max Activating Token Index: 506

Text #16

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0420. Min Act: -0.0004

Data Index: 450078 (C4 (Web Text))

Max Activating Token Index: 388

Text #17

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0417. Min Act: -0.0003

Data Index: 695230 (C4 (Web Text))

Max Activating Token Index: 84

Text #18

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0408. Min Act: -0.0003

Data Index: 735980 (C4 (Web Text))

Max Activating Token Index: 412

Text #19

Max Range: 0.0463. Min Range: -0.0463

Max Act: 0.0415. Min Act: -0.0004

Data Index: 112777 (C4 (Web Text))

Max Activating Token Index: 398