Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1559 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0243. Min Act: -0.0006

Data Index: 1322229 (C4 (Web Text))

Max Activating Token Index: 181

Text #1

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0184. Min Act: -0.0005

Data Index: 1414727 (Python Code)

Max Activating Token Index: 693

Text #2

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0173. Min Act: -0.0006

Data Index: 1189949 (C4 (Web Text))

Max Activating Token Index: 387

Text #3

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0176. Min Act: -0.0005

Data Index: 1382908 (Python Code)

Max Activating Token Index: 355

Text #4

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0174. Min Act: -0.0006

Data Index: 847727 (C4 (Web Text))

Max Activating Token Index: 1007

Text #5

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0170. Min Act: -0.0006

Data Index: 107614 (C4 (Web Text))

Max Activating Token Index: 693

Text #6

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0178. Min Act: -0.0005

Data Index: 1489700 (Python Code)

Max Activating Token Index: 409

Text #7

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0173. Min Act: -0.0006

Data Index: 1348329 (C4 (Web Text))

Max Activating Token Index: 419

Text #8

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0170. Min Act: -0.0006

Data Index: 1606999 (Python Code)

Max Activating Token Index: 514

Text #9

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0169. Min Act: -0.0005

Data Index: 643625 (C4 (Web Text))

Max Activating Token Index: 73

Text #10

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0168. Min Act: -0.0006

Data Index: 1523281 (Python Code)

Max Activating Token Index: 94

Text #11

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0167. Min Act: -0.0006

Data Index: 546752 (C4 (Web Text))

Max Activating Token Index: 168

Text #12

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0163. Min Act: -0.0006

Data Index: 11638 (C4 (Web Text))

Max Activating Token Index: 1000

Text #13

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0165. Min Act: -0.0006

Data Index: 1054322 (C4 (Web Text))

Max Activating Token Index: 845

Text #14

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0160. Min Act: -0.0006

Data Index: 1568559 (Python Code)

Max Activating Token Index: 745

Text #15

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0168. Min Act: -0.0006

Data Index: 706992 (C4 (Web Text))

Max Activating Token Index: 236

Text #16

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0167. Min Act: -0.0006

Data Index: 595319 (C4 (Web Text))

Max Activating Token Index: 1019

Text #17

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0167. Min Act: -0.0006

Data Index: 128364 (C4 (Web Text))

Max Activating Token Index: 260

Text #18

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0162. Min Act: -0.0006

Data Index: 1088909 (C4 (Web Text))

Max Activating Token Index: 98

Text #19

Max Range: 0.0243. Min Range: -0.0243

Max Act: 0.0166. Min Act: -0.0005

Data Index: 1616396 (Python Code)

Max Activating Token Index: 222