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

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

Neuron 5482 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0454. Min Act: -0.0001

Data Index: 1087875 (C4 (Web Text))

Max Activating Token Index: 358

Text #1

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0370. Min Act: -0.0001

Data Index: 475338 (C4 (Web Text))

Max Activating Token Index: 742

Text #2

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0369. Min Act: -0.0001

Data Index: 1265671 (C4 (Web Text))

Max Activating Token Index: 240

Text #3

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0372. Min Act: -0.0001

Data Index: 510516 (C4 (Web Text))

Max Activating Token Index: 290

Text #4

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0391. Min Act: -0.0001

Data Index: 1315798 (C4 (Web Text))

Max Activating Token Index: 218

Text #5

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0369. Min Act: -0.0001

Data Index: 455971 (C4 (Web Text))

Max Activating Token Index: 604

Text #6

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0382. Min Act: -0.0002

Data Index: 931459 (C4 (Web Text))

Max Activating Token Index: 30

Text #7

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0382. Min Act: -0.0001

Data Index: 580660 (C4 (Web Text))

Max Activating Token Index: 402

Text #8

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0356. Min Act: -0.0001

Data Index: 829922 (C4 (Web Text))

Max Activating Token Index: 300

Text #9

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0392. Min Act: -0.0001

Data Index: 1209210 (C4 (Web Text))

Max Activating Token Index: 216

Text #10

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0420. Min Act: -0.0001

Data Index: 453902 (C4 (Web Text))

Max Activating Token Index: 117

Text #11

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0406. Min Act: -0.0001

Data Index: 599708 (C4 (Web Text))

Max Activating Token Index: 340

Text #12

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0382. Min Act: -0.0001

Data Index: 181670 (C4 (Web Text))

Max Activating Token Index: 485

Text #13

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0415. Min Act: -0.0001

Data Index: 164708 (C4 (Web Text))

Max Activating Token Index: 274

Text #14

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0383. Min Act: -0.0001

Data Index: 1258018 (C4 (Web Text))

Max Activating Token Index: 384

Text #15

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0367. Min Act: -0.0001

Data Index: 63525 (C4 (Web Text))

Max Activating Token Index: 98

Text #16

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0328. Min Act: -0.0001

Data Index: 1280496 (C4 (Web Text))

Max Activating Token Index: 826

Text #17

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0395. Min Act: -0.0001

Data Index: 1646407 (Python Code)

Max Activating Token Index: 4

Text #18

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0341. Min Act: -0.0001

Data Index: 1177098 (C4 (Web Text))

Max Activating Token Index: 162

Text #19

Max Range: 0.0454. Min Range: -0.0454

Max Act: 0.0350. Min Act: -0.0001

Data Index: 342827 (C4 (Web Text))

Max Activating Token Index: 100