Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 546 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1558. Min Act: -0.0003

Data Index: 719688 (C4 (Web Text))

Max Activating Token Index: 646

Text #1

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1531. Min Act: -0.0003

Data Index: 222925 (C4 (Web Text))

Max Activating Token Index: 192

Text #2

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1496. Min Act: -0.0003

Data Index: 1086319 (C4 (Web Text))

Max Activating Token Index: 348

Text #3

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1378. Min Act: -0.0003

Data Index: 1437806 (Python Code)

Max Activating Token Index: 516

Text #4

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1377. Min Act: -0.0003

Data Index: 932672 (C4 (Web Text))

Max Activating Token Index: 779

Text #5

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1317. Min Act: -0.0002

Data Index: 588271 (C4 (Web Text))

Max Activating Token Index: 1006

Text #6

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1339. Min Act: -0.0003

Data Index: 1582097 (Python Code)

Max Activating Token Index: 586

Text #7

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1306. Min Act: -0.0002

Data Index: 688230 (C4 (Web Text))

Max Activating Token Index: 1010

Text #8

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1288. Min Act: -0.0003

Data Index: 1528501 (Python Code)

Max Activating Token Index: 867

Text #9

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1335. Min Act: -0.0003

Data Index: 1528574 (Python Code)

Max Activating Token Index: 803

Text #10

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1281. Min Act: -0.0003

Data Index: 749778 (C4 (Web Text))

Max Activating Token Index: 110

Text #11

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1359. Min Act: -0.0003

Data Index: 739275 (C4 (Web Text))

Max Activating Token Index: 961

Text #12

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1243. Min Act: -0.0002

Data Index: 1420175 (Python Code)

Max Activating Token Index: 435

Text #13

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1282. Min Act: -0.0002

Data Index: 1525844 (Python Code)

Max Activating Token Index: 305

Text #14

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1271. Min Act: -0.0003

Data Index: 1591141 (Python Code)

Max Activating Token Index: 462

Text #15

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1237. Min Act: -0.0003

Data Index: 1452315 (Python Code)

Max Activating Token Index: 977

Text #16

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1272. Min Act: -0.0002

Data Index: 697188 (C4 (Web Text))

Max Activating Token Index: 809

Text #17

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1091. Min Act: -0.0003

Data Index: 881922 (C4 (Web Text))

Max Activating Token Index: 835

Text #18

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1201. Min Act: -0.0003

Data Index: 1296059 (C4 (Web Text))

Max Activating Token Index: 318

Text #19

Max Range: 0.1558. Min Range: -0.1558

Max Act: 0.1209. Min Act: -0.0003

Data Index: 1611664 (Python Code)

Max Activating Token Index: 369