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

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

Neuron 3192 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: 0.0253. Min Range: -0.0253

Max Act: 0.0253. Min Act: -0.0001

Data Index: 980306 (C4 (Web Text))

Max Activating Token Index: 829

Text #1

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0238. Min Act: -0.0001

Data Index: 1022071 (C4 (Web Text))

Max Activating Token Index: 81

Text #2

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0241. Min Act: -0.0001

Data Index: 681831 (C4 (Web Text))

Max Activating Token Index: 923

Text #3

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0240. Min Act: -0.0001

Data Index: 981126 (C4 (Web Text))

Max Activating Token Index: 470

Text #4

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0220. Min Act: -0.0001

Data Index: 520320 (C4 (Web Text))

Max Activating Token Index: 649

Text #5

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0222. Min Act: -0.0001

Data Index: 164239 (C4 (Web Text))

Max Activating Token Index: 71

Text #6

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0207. Min Act: -0.0001

Data Index: 832275 (C4 (Web Text))

Max Activating Token Index: 757

Text #7

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0205. Min Act: -0.0001

Data Index: 33647 (C4 (Web Text))

Max Activating Token Index: 534

Text #8

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0228. Min Act: -0.0001

Data Index: 33354 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0218. Min Act: -0.0001

Data Index: 1502865 (Python Code)

Max Activating Token Index: 32

Text #10

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0243. Min Act: -0.0001

Data Index: 417985 (C4 (Web Text))

Max Activating Token Index: 597

Text #11

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0196. Min Act: -0.0001

Data Index: 116155 (C4 (Web Text))

Max Activating Token Index: 196

Text #12

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0195. Min Act: -0.0001

Data Index: 387332 (C4 (Web Text))

Max Activating Token Index: 901

Text #13

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0202. Min Act: -0.0001

Data Index: 951217 (C4 (Web Text))

Max Activating Token Index: 725

Text #14

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0193. Min Act: -0.0001

Data Index: 389740 (C4 (Web Text))

Max Activating Token Index: 71

Text #15

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0202. Min Act: -0.0001

Data Index: 293572 (C4 (Web Text))

Max Activating Token Index: 497

Text #16

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0192. Min Act: -0.0001

Data Index: 828035 (C4 (Web Text))

Max Activating Token Index: 574

Text #17

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0199. Min Act: -0.0001

Data Index: 1550591 (Python Code)

Max Activating Token Index: 9

Text #18

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0187. Min Act: -0.0001

Data Index: 1080283 (C4 (Web Text))

Max Activating Token Index: 64

Text #19

Max Range: 0.0253. Min Range: -0.0253

Max Act: 0.0192. Min Act: -0.0001

Data Index: 566752 (C4 (Web Text))

Max Activating Token Index: 93