Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 826 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.2096. Min Act: -0.0001

Data Index: 909107 (C4 (Web Text))

Max Activating Token Index: 503

Text #1

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1422. Min Act: -0.0002

Data Index: 1393215 (Python Code)

Max Activating Token Index: 545

Text #2

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1397. Min Act: -0.0002

Data Index: 991147 (C4 (Web Text))

Max Activating Token Index: 512

Text #3

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1491. Min Act: -0.0002

Data Index: 1328933 (C4 (Web Text))

Max Activating Token Index: 239

Text #4

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1391. Min Act: -0.0002

Data Index: 644102 (C4 (Web Text))

Max Activating Token Index: 345

Text #5

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1278. Min Act: -0.0002

Data Index: 723948 (C4 (Web Text))

Max Activating Token Index: 549

Text #6

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1274. Min Act: -0.0001

Data Index: 848624 (C4 (Web Text))

Max Activating Token Index: 830

Text #7

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1371. Min Act: -0.0002

Data Index: 1480262 (Python Code)

Max Activating Token Index: 427

Text #8

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1280. Min Act: -0.0002

Data Index: 8877 (C4 (Web Text))

Max Activating Token Index: 584

Text #9

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1303. Min Act: -0.0002

Data Index: 1189290 (C4 (Web Text))

Max Activating Token Index: 273

Text #10

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1251. Min Act: -0.0002

Data Index: 1612129 (Python Code)

Max Activating Token Index: 599

Text #11

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1282. Min Act: -0.0002

Data Index: 1494648 (Python Code)

Max Activating Token Index: 936

Text #12

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1242. Min Act: -0.0002

Data Index: 1003038 (C4 (Web Text))

Max Activating Token Index: 604

Text #13

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1284. Min Act: -0.0002

Data Index: 1610751 (Python Code)

Max Activating Token Index: 614

Text #14

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1259. Min Act: -0.0002

Data Index: 1084621 (C4 (Web Text))

Max Activating Token Index: 975

Text #15

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1269. Min Act: -0.0002

Data Index: 1449703 (Python Code)

Max Activating Token Index: 174

Text #16

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1212. Min Act: -0.0002

Data Index: 1413288 (Python Code)

Max Activating Token Index: 292

Text #17

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1283. Min Act: -0.0002

Data Index: 1596552 (Python Code)

Max Activating Token Index: 229

Text #18

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1104. Min Act: -0.0002

Data Index: 1004941 (C4 (Web Text))

Max Activating Token Index: 489

Text #19

Max Range: 0.2096. Min Range: -0.2096

Max Act: 0.1242. Min Act: -0.0002

Data Index: 1588143 (Python Code)

Max Activating Token Index: 347