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

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

Neuron 2549 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: 0.0426. Min Range: -0.0426

Max Act: 0.0426. Min Act: -0.0001

Data Index: 1299318 (C4 (Web Text))

Max Activating Token Index: 67

Text #1

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0392. Min Act: -0.0001

Data Index: 1495463 (Python Code)

Max Activating Token Index: 246

Text #2

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0349. Min Act: -0.0001

Data Index: 191710 (C4 (Web Text))

Max Activating Token Index: 482

Text #3

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0357. Min Act: -0.0001

Data Index: 1208802 (C4 (Web Text))

Max Activating Token Index: 508

Text #4

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0321. Min Act: -0.0001

Data Index: 114726 (C4 (Web Text))

Max Activating Token Index: 456

Text #5

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0325. Min Act: -0.0001

Data Index: 391619 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0294. Min Act: -0.0001

Data Index: 882509 (C4 (Web Text))

Max Activating Token Index: 401

Text #7

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0301. Min Act: -0.0002

Data Index: 1386575 (Python Code)

Max Activating Token Index: 852

Text #8

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0311. Min Act: -0.0001

Data Index: 698914 (C4 (Web Text))

Max Activating Token Index: 401

Text #9

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0317. Min Act: -0.0001

Data Index: 1249658 (C4 (Web Text))

Max Activating Token Index: 213

Text #10

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0268. Min Act: -0.0001

Data Index: 841556 (C4 (Web Text))

Max Activating Token Index: 828

Text #11

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0323. Min Act: -0.0001

Data Index: 1258187 (C4 (Web Text))

Max Activating Token Index: 852

Text #12

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0270. Min Act: -0.0001

Data Index: 1057971 (C4 (Web Text))

Max Activating Token Index: 89

Text #13

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0293. Min Act: -0.0001

Data Index: 1648344 (Python Code)

Max Activating Token Index: 590

Text #14

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0341. Min Act: -0.0001

Data Index: 676369 (C4 (Web Text))

Max Activating Token Index: 413

Text #15

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0268. Min Act: -0.0001

Data Index: 406064 (C4 (Web Text))

Max Activating Token Index: 104

Text #16

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0275. Min Act: -0.0001

Data Index: 1604548 (Python Code)

Max Activating Token Index: 887

Text #17

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0285. Min Act: -0.0001

Data Index: 363865 (C4 (Web Text))

Max Activating Token Index: 755

Text #18

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0248. Min Act: -0.0001

Data Index: 218260 (C4 (Web Text))

Max Activating Token Index: 958

Text #19

Max Range: 0.0426. Min Range: -0.0426

Max Act: 0.0252. Min Act: -0.0001

Data Index: 1039329 (C4 (Web Text))

Max Activating Token Index: 725