Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2631 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0283. Min Act: -0.0001

Data Index: 929464 (C4 (Web Text))

Max Activating Token Index: 990

Text #1

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0280. Min Act: -0.0001

Data Index: 73731 (C4 (Web Text))

Max Activating Token Index: 318

Text #2

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0292. Min Act: -0.0001

Data Index: 47458 (C4 (Web Text))

Max Activating Token Index: 653

Text #3

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0270. Min Act: -0.0001

Data Index: 443346 (C4 (Web Text))

Max Activating Token Index: 807

Text #4

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0254. Min Act: -0.0001

Data Index: 1595549 (Python Code)

Max Activating Token Index: 1016

Text #5

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0266. Min Act: -0.0001

Data Index: 512526 (C4 (Web Text))

Max Activating Token Index: 886

Text #6

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0257. Min Act: -0.0001

Data Index: 70604 (C4 (Web Text))

Max Activating Token Index: 339

Text #7

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0266. Min Act: -0.0001

Data Index: 360077 (C4 (Web Text))

Max Activating Token Index: 501

Text #8

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0265. Min Act: -0.0001

Data Index: 63459 (C4 (Web Text))

Max Activating Token Index: 870

Text #9

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0252. Min Act: -0.0001

Data Index: 1433459 (Python Code)

Max Activating Token Index: 644

Text #10

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0280. Min Act: -0.0001

Data Index: 444414 (C4 (Web Text))

Max Activating Token Index: 835

Text #11

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0252. Min Act: -0.0001

Data Index: 528963 (C4 (Web Text))

Max Activating Token Index: 959

Text #12

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0239. Min Act: -0.0001

Data Index: 845195 (C4 (Web Text))

Max Activating Token Index: 893

Text #13

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0227. Min Act: -0.0001

Data Index: 344353 (C4 (Web Text))

Max Activating Token Index: 881

Text #14

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0254. Min Act: -0.0001

Data Index: 123102 (C4 (Web Text))

Max Activating Token Index: 649

Text #15

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0240. Min Act: -0.0001

Data Index: 1564298 (Python Code)

Max Activating Token Index: 326

Text #16

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0263. Min Act: -0.0001

Data Index: 285150 (C4 (Web Text))

Max Activating Token Index: 314

Text #17

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0222. Min Act: -0.0001

Data Index: 1445732 (Python Code)

Max Activating Token Index: 699

Text #18

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0250. Min Act: -0.0001

Data Index: 1454465 (Python Code)

Max Activating Token Index: 521

Text #19

Max Range: 0.0292. Min Range: -0.0292

Max Act: 0.0242. Min Act: -0.0001

Data Index: 543536 (C4 (Web Text))

Max Activating Token Index: 683