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

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

Neuron 3128 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1601. Min Act: -0.0002

Data Index: 1383053 (Python Code)

Max Activating Token Index: 963

Text #1

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1483. Min Act: -0.0002

Data Index: 1635092 (Python Code)

Max Activating Token Index: 1018

Text #2

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1372. Min Act: -0.0002

Data Index: 2735 (C4 (Web Text))

Max Activating Token Index: 413

Text #3

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1340. Min Act: -0.0002

Data Index: 528953 (C4 (Web Text))

Max Activating Token Index: 441

Text #4

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1393. Min Act: -0.0002

Data Index: 574155 (C4 (Web Text))

Max Activating Token Index: 870

Text #5

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1335. Min Act: -0.0002

Data Index: 187040 (C4 (Web Text))

Max Activating Token Index: 411

Text #6

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1279. Min Act: -0.0002

Data Index: 504949 (C4 (Web Text))

Max Activating Token Index: 486

Text #7

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1215. Min Act: -0.0002

Data Index: 488439 (C4 (Web Text))

Max Activating Token Index: 432

Text #8

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1344. Min Act: -0.0002

Data Index: 28581 (C4 (Web Text))

Max Activating Token Index: 794

Text #9

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1241. Min Act: -0.0002

Data Index: 921613 (C4 (Web Text))

Max Activating Token Index: 323

Text #10

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1329. Min Act: -0.0002

Data Index: 1319725 (C4 (Web Text))

Max Activating Token Index: 431

Text #11

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1242. Min Act: -0.0002

Data Index: 1559031 (Python Code)

Max Activating Token Index: 824

Text #12

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1234. Min Act: -0.0002

Data Index: 721929 (C4 (Web Text))

Max Activating Token Index: 165

Text #13

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1254. Min Act: -0.0002

Data Index: 99388 (C4 (Web Text))

Max Activating Token Index: 325

Text #14

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1127. Min Act: -0.0002

Data Index: 133903 (C4 (Web Text))

Max Activating Token Index: 356

Text #15

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1156. Min Act: -0.0002

Data Index: 1379845 (Python Code)

Max Activating Token Index: 954

Text #16

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1241. Min Act: -0.0002

Data Index: 968648 (C4 (Web Text))

Max Activating Token Index: 725

Text #17

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1225. Min Act: -0.0002

Data Index: 1160125 (C4 (Web Text))

Max Activating Token Index: 336

Text #18

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1173. Min Act: -0.0002

Data Index: 765474 (C4 (Web Text))

Max Activating Token Index: 531

Text #19

Max Range: 0.1601. Min Range: -0.1601

Max Act: 0.1232. Min Act: -0.0002

Data Index: 1322363 (C4 (Web Text))

Max Activating Token Index: 401