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

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

Neuron 3090 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0415. Min Act: -0.0001

Data Index: 400669 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0329. Min Act: -0.0001

Data Index: 1197294 (C4 (Web Text))

Max Activating Token Index: 216

Text #2

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0382. Min Act: -0.0001

Data Index: 1163267 (C4 (Web Text))

Max Activating Token Index: 201

Text #3

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0353. Min Act: -0.0000

Data Index: 177472 (C4 (Web Text))

Max Activating Token Index: 156

Text #4

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0339. Min Act: -0.0001

Data Index: 351807 (C4 (Web Text))

Max Activating Token Index: 387

Text #5

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0326. Min Act: -0.0001

Data Index: 570722 (C4 (Web Text))

Max Activating Token Index: 1020

Text #6

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0324. Min Act: -0.0001

Data Index: 479880 (C4 (Web Text))

Max Activating Token Index: 701

Text #7

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0340. Min Act: -0.0001

Data Index: 944249 (C4 (Web Text))

Max Activating Token Index: 801

Text #8

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0317. Min Act: -0.0001

Data Index: 741760 (C4 (Web Text))

Max Activating Token Index: 498

Text #9

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0310. Min Act: -0.0001

Data Index: 663192 (C4 (Web Text))

Max Activating Token Index: 117

Text #10

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0290. Min Act: -0.0001

Data Index: 1356608 (C4 (Web Text))

Max Activating Token Index: 826

Text #11

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0316. Min Act: -0.0001

Data Index: 500132 (C4 (Web Text))

Max Activating Token Index: 698

Text #12

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0311. Min Act: -0.0001

Data Index: 1425595 (Python Code)

Max Activating Token Index: 753

Text #13

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0304. Min Act: -0.0001

Data Index: 227244 (C4 (Web Text))

Max Activating Token Index: 529

Text #14

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0304. Min Act: -0.0001

Data Index: 1197403 (C4 (Web Text))

Max Activating Token Index: 128

Text #15

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0311. Min Act: -0.0001

Data Index: 1229061 (C4 (Web Text))

Max Activating Token Index: 99

Text #16

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0299. Min Act: -0.0001

Data Index: 131110 (C4 (Web Text))

Max Activating Token Index: 139

Text #17

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0303. Min Act: -0.0000

Data Index: 334496 (C4 (Web Text))

Max Activating Token Index: 655

Text #18

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0309. Min Act: -0.0001

Data Index: 630630 (C4 (Web Text))

Max Activating Token Index: 912

Text #19

Max Range: 0.0415. Min Range: -0.0415

Max Act: 0.0285. Min Act: -0.0001

Data Index: 1312328 (C4 (Web Text))

Max Activating Token Index: 236