Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4715 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0380. Min Act: -0.0001

Data Index: 323271 (C4 (Web Text))

Max Activating Token Index: 284

Text #1

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0377. Min Act: -0.0001

Data Index: 1073203 (C4 (Web Text))

Max Activating Token Index: 451

Text #2

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0328. Min Act: -0.0001

Data Index: 105300 (C4 (Web Text))

Max Activating Token Index: 264

Text #3

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0309. Min Act: -0.0001

Data Index: 492525 (C4 (Web Text))

Max Activating Token Index: 953

Text #4

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0306. Min Act: -0.0001

Data Index: 515000 (C4 (Web Text))

Max Activating Token Index: 800

Text #5

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0309. Min Act: -0.0001

Data Index: 976204 (C4 (Web Text))

Max Activating Token Index: 753

Text #6

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0307. Min Act: -0.0001

Data Index: 1290741 (C4 (Web Text))

Max Activating Token Index: 148

Text #7

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0273. Min Act: -0.0001

Data Index: 1036731 (C4 (Web Text))

Max Activating Token Index: 340

Text #8

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0293. Min Act: -0.0001

Data Index: 1619895 (Python Code)

Max Activating Token Index: 786

Text #9

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0288. Min Act: -0.0001

Data Index: 689880 (C4 (Web Text))

Max Activating Token Index: 904

Text #10

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0263. Min Act: -0.0001

Data Index: 1152907 (C4 (Web Text))

Max Activating Token Index: 611

Text #11

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0277. Min Act: -0.0001

Data Index: 103232 (C4 (Web Text))

Max Activating Token Index: 148

Text #12

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0272. Min Act: -0.0001

Data Index: 308742 (C4 (Web Text))

Max Activating Token Index: 824

Text #13

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0301. Min Act: -0.0001

Data Index: 1206744 (C4 (Web Text))

Max Activating Token Index: 732

Text #14

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0265. Min Act: -0.0001

Data Index: 1100743 (C4 (Web Text))

Max Activating Token Index: 352

Text #15

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0265. Min Act: -0.0001

Data Index: 1037293 (C4 (Web Text))

Max Activating Token Index: 416

Text #16

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0284. Min Act: -0.0001

Data Index: 1270060 (C4 (Web Text))

Max Activating Token Index: 765

Text #17

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0261. Min Act: -0.0001

Data Index: 774140 (C4 (Web Text))

Max Activating Token Index: 665

Text #18

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0251. Min Act: -0.0001

Data Index: 903196 (C4 (Web Text))

Max Activating Token Index: 1004

Text #19

Max Range: 0.0380. Min Range: -0.0380

Max Act: 0.0262. Min Act: -0.0001

Data Index: 627979 (C4 (Web Text))

Max Activating Token Index: 580