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

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

Neuron 4230 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0275. Min Act: -0.0000

Data Index: 823387 (C4 (Web Text))

Max Activating Token Index: 552

Text #1

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0258. Min Act: -0.0000

Data Index: 642257 (C4 (Web Text))

Max Activating Token Index: 420

Text #2

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0249. Min Act: -0.0000

Data Index: 1018016 (C4 (Web Text))

Max Activating Token Index: 664

Text #3

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0243. Min Act: -0.0000

Data Index: 819361 (C4 (Web Text))

Max Activating Token Index: 139

Text #4

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0250. Min Act: -0.0000

Data Index: 1036294 (C4 (Web Text))

Max Activating Token Index: 709

Text #5

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0230. Min Act: -0.0000

Data Index: 366238 (C4 (Web Text))

Max Activating Token Index: 634

Text #6

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0210. Min Act: -0.0000

Data Index: 637100 (C4 (Web Text))

Max Activating Token Index: 571

Text #7

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0194. Min Act: -0.0000

Data Index: 1619034 (Python Code)

Max Activating Token Index: 756

Text #8

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0216. Min Act: -0.0000

Data Index: 966303 (C4 (Web Text))

Max Activating Token Index: 945

Text #9

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0204. Min Act: -0.0000

Data Index: 494329 (C4 (Web Text))

Max Activating Token Index: 855

Text #10

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0188. Min Act: -0.0000

Data Index: 1028512 (C4 (Web Text))

Max Activating Token Index: 708

Text #11

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0201. Min Act: -0.0000

Data Index: 1121638 (C4 (Web Text))

Max Activating Token Index: 523

Text #12

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0191. Min Act: -0.0000

Data Index: 1465557 (Python Code)

Max Activating Token Index: 396

Text #13

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0190. Min Act: -0.0000

Data Index: 962245 (C4 (Web Text))

Max Activating Token Index: 324

Text #14

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0190. Min Act: -0.0000

Data Index: 1540097 (Python Code)

Max Activating Token Index: 881

Text #15

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0169. Min Act: -0.0000

Data Index: 176717 (C4 (Web Text))

Max Activating Token Index: 189

Text #16

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0182. Min Act: -0.0000

Data Index: 827780 (C4 (Web Text))

Max Activating Token Index: 150

Text #17

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0186. Min Act: -0.0000

Data Index: 1081469 (C4 (Web Text))

Max Activating Token Index: 907

Text #18

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0188. Min Act: -0.0000

Data Index: 505008 (C4 (Web Text))

Max Activating Token Index: 909

Text #19

Max Range: 0.0275. Min Range: -0.0275

Max Act: 0.0178. Min Act: -0.0000

Data Index: 117381 (C4 (Web Text))

Max Activating Token Index: 753