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

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

Neuron 3456 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0390. Min Act: -0.0003

Data Index: 364554 (C4 (Web Text))

Max Activating Token Index: 188

Text #1

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0362. Min Act: -0.0003

Data Index: 877630 (C4 (Web Text))

Max Activating Token Index: 971

Text #2

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0342. Min Act: -0.0003

Data Index: 1358361 (C4 (Web Text))

Max Activating Token Index: 196

Text #3

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0348. Min Act: -0.0003

Data Index: 43638 (C4 (Web Text))

Max Activating Token Index: 187

Text #4

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0324. Min Act: -0.0003

Data Index: 1427239 (Python Code)

Max Activating Token Index: 319

Text #5

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0321. Min Act: -0.0003

Data Index: 66314 (C4 (Web Text))

Max Activating Token Index: 395

Text #6

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0312. Min Act: -0.0003

Data Index: 869302 (C4 (Web Text))

Max Activating Token Index: 206

Text #7

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0316. Min Act: -0.0003

Data Index: 1081359 (C4 (Web Text))

Max Activating Token Index: 959

Text #8

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0298. Min Act: -0.0003

Data Index: 1277277 (C4 (Web Text))

Max Activating Token Index: 88

Text #9

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0308. Min Act: -0.0002

Data Index: 406270 (C4 (Web Text))

Max Activating Token Index: 867

Text #10

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0295. Min Act: -0.0003

Data Index: 1346416 (C4 (Web Text))

Max Activating Token Index: 367

Text #11

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0298. Min Act: -0.0003

Data Index: 533106 (C4 (Web Text))

Max Activating Token Index: 445

Text #12

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0288. Min Act: -0.0003

Data Index: 210797 (C4 (Web Text))

Max Activating Token Index: 250

Text #13

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0289. Min Act: -0.0003

Data Index: 468713 (C4 (Web Text))

Max Activating Token Index: 404

Text #14

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0284. Min Act: -0.0003

Data Index: 1200382 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0295. Min Act: -0.0002

Data Index: 192284 (C4 (Web Text))

Max Activating Token Index: 34

Text #16

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0290. Min Act: -0.0003

Data Index: 1004774 (C4 (Web Text))

Max Activating Token Index: 193

Text #17

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0285. Min Act: -0.0003

Data Index: 106762 (C4 (Web Text))

Max Activating Token Index: 518

Text #18

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0294. Min Act: -0.0003

Data Index: 1528989 (Python Code)

Max Activating Token Index: 201

Text #19

Max Range: 0.0390. Min Range: -0.0390

Max Act: 0.0286. Min Act: -0.0003

Data Index: 158833 (C4 (Web Text))

Max Activating Token Index: 561