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

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

Neuron 1380 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: 0.0402. Min Range: -0.0402

Max Act: 0.0386. Min Act: -0.0001

Data Index: 860749 (C4 (Web Text))

Max Activating Token Index: 39

Text #1

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0402. Min Act: -0.0001

Data Index: 1484979 (Python Code)

Max Activating Token Index: 489

Text #2

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0378. Min Act: -0.0001

Data Index: 771757 (C4 (Web Text))

Max Activating Token Index: 247

Text #3

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0343. Min Act: -0.0001

Data Index: 756849 (C4 (Web Text))

Max Activating Token Index: 336

Text #4

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0207. Min Act: -0.0001

Data Index: 615051 (C4 (Web Text))

Max Activating Token Index: 348

Text #5

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0350. Min Act: -0.0001

Data Index: 77948 (C4 (Web Text))

Max Activating Token Index: 489

Text #6

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0339. Min Act: -0.0001

Data Index: 139572 (C4 (Web Text))

Max Activating Token Index: 216

Text #7

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0341. Min Act: -0.0001

Data Index: 1557024 (Python Code)

Max Activating Token Index: 695

Text #8

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0385. Min Act: -0.0001

Data Index: 1283828 (C4 (Web Text))

Max Activating Token Index: 215

Text #9

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0342. Min Act: -0.0001

Data Index: 985169 (C4 (Web Text))

Max Activating Token Index: 203

Text #10

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0326. Min Act: -0.0001

Data Index: 1192218 (C4 (Web Text))

Max Activating Token Index: 2

Text #11

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0326. Min Act: -0.0001

Data Index: 227815 (C4 (Web Text))

Max Activating Token Index: 2

Text #12

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0321. Min Act: -0.0001

Data Index: 1532349 (Python Code)

Max Activating Token Index: 8

Text #13

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0321. Min Act: -0.0001

Data Index: 847473 (C4 (Web Text))

Max Activating Token Index: 8

Text #14

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0393. Min Act: -0.0001

Data Index: 1037781 (C4 (Web Text))

Max Activating Token Index: 311

Text #15

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0326. Min Act: -0.0001

Data Index: 1383483 (Python Code)

Max Activating Token Index: 13

Text #16

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0298. Min Act: -0.0001

Data Index: 1242213 (C4 (Web Text))

Max Activating Token Index: 309

Text #17

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0305. Min Act: -0.0001

Data Index: 1209355 (C4 (Web Text))

Max Activating Token Index: 130

Text #18

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0318. Min Act: -0.0001

Data Index: 1635978 (Python Code)

Max Activating Token Index: 272

Text #19

Max Range: 0.0402. Min Range: -0.0402

Max Act: 0.0357. Min Act: -0.0001

Data Index: 431550 (C4 (Web Text))

Max Activating Token Index: 871