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

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

Neuron 4344 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0207. Min Act: -0.0003

Data Index: 802856 (C4 (Web Text))

Max Activating Token Index: 469

Text #1

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0196. Min Act: -0.0003

Data Index: 575404 (C4 (Web Text))

Max Activating Token Index: 813

Text #2

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0159. Min Act: -0.0003

Data Index: 909642 (C4 (Web Text))

Max Activating Token Index: 194

Text #3

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0179. Min Act: -0.0003

Data Index: 1511908 (Python Code)

Max Activating Token Index: 988

Text #4

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0165. Min Act: -0.0003

Data Index: 1450632 (Python Code)

Max Activating Token Index: 342

Text #5

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0164. Min Act: -0.0002

Data Index: 858056 (C4 (Web Text))

Max Activating Token Index: 342

Text #6

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0163. Min Act: -0.0002

Data Index: 1012879 (C4 (Web Text))

Max Activating Token Index: 451

Text #7

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0158. Min Act: -0.0003

Data Index: 798152 (C4 (Web Text))

Max Activating Token Index: 946

Text #8

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0153. Min Act: -0.0002

Data Index: 711792 (C4 (Web Text))

Max Activating Token Index: 466

Text #9

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0156. Min Act: -0.0003

Data Index: 1409789 (Python Code)

Max Activating Token Index: 259

Text #10

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0153. Min Act: -0.0003

Data Index: 1385276 (Python Code)

Max Activating Token Index: 534

Text #11

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0148. Min Act: -0.0003

Data Index: 1469285 (Python Code)

Max Activating Token Index: 303

Text #12

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0156. Min Act: -0.0003

Data Index: 845710 (C4 (Web Text))

Max Activating Token Index: 555

Text #13

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0144. Min Act: -0.0003

Data Index: 720151 (C4 (Web Text))

Max Activating Token Index: 326

Text #14

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0146. Min Act: -0.0003

Data Index: 102417 (C4 (Web Text))

Max Activating Token Index: 21

Text #15

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0147. Min Act: -0.0003

Data Index: 1246809 (C4 (Web Text))

Max Activating Token Index: 415

Text #16

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0143. Min Act: -0.0003

Data Index: 1105134 (C4 (Web Text))

Max Activating Token Index: 393

Text #17

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0149. Min Act: -0.0003

Data Index: 1601114 (Python Code)

Max Activating Token Index: 707

Text #18

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0139. Min Act: -0.0002

Data Index: 1373708 (Python Code)

Max Activating Token Index: 153

Text #19

Max Range: 0.0207. Min Range: -0.0207

Max Act: 0.0135. Min Act: -0.0002

Data Index: 998410 (C4 (Web Text))

Max Activating Token Index: 245