Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5483 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.4153. Min Range: -4.4153

Max Act: 4.4153. Min Act: -0.0000

Data Index: 425759 (C4 (Web Text))

Max Activating Token Index: 352

Text #1

Max Range: 4.4153. Min Range: -4.4153

Max Act: 4.4124. Min Act: -0.0000

Data Index: 981282 (C4 (Web Text))

Max Activating Token Index: 342

Text #2

Max Range: 4.4153. Min Range: -4.4153

Max Act: 3.8449. Min Act: -0.0000

Data Index: 1158480 (C4 (Web Text))

Max Activating Token Index: 94

Text #3

Max Range: 4.4153. Min Range: -4.4153

Max Act: 3.5532. Min Act: -0.0000

Data Index: 274964 (C4 (Web Text))

Max Activating Token Index: 559

Text #4

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.9908. Min Act: -0.0000

Data Index: 216857 (C4 (Web Text))

Max Activating Token Index: 734

Text #5

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.7134. Min Act: -0.0000

Data Index: 627088 (C4 (Web Text))

Max Activating Token Index: 8

Text #6

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.6246. Min Act: -0.0000

Data Index: 569059 (C4 (Web Text))

Max Activating Token Index: 503

Text #7

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.4301. Min Act: -0.0000

Data Index: 1514256 (Python Code)

Max Activating Token Index: 266

Text #8

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.4139. Min Act: -0.0000

Data Index: 1480758 (Python Code)

Max Activating Token Index: 928

Text #9

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.1566. Min Act: -0.0000

Data Index: 1649435 (Python Code)

Max Activating Token Index: 688

Text #10

Max Range: 4.4153. Min Range: -4.4153

Max Act: 1.9813. Min Act: -0.0000

Data Index: 263907 (C4 (Web Text))

Max Activating Token Index: 927

Text #11

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.3232. Min Act: -0.0000

Data Index: 663306 (C4 (Web Text))

Max Activating Token Index: 830

Text #12

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.0746. Min Act: -0.0000

Data Index: 1429738 (Python Code)

Max Activating Token Index: 507

Text #13

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.1469. Min Act: -0.0000

Data Index: 157413 (C4 (Web Text))

Max Activating Token Index: 943

Text #14

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.1219. Min Act: -0.0000

Data Index: 844476 (C4 (Web Text))

Max Activating Token Index: 108

Text #15

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.1075. Min Act: -0.0000

Data Index: 1093048 (C4 (Web Text))

Max Activating Token Index: 902

Text #16

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.3081. Min Act: -0.0000

Data Index: 774139 (C4 (Web Text))

Max Activating Token Index: 231

Text #17

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.0735. Min Act: -0.0000

Data Index: 1056549 (C4 (Web Text))

Max Activating Token Index: 293

Text #18

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.0056. Min Act: -0.0000

Data Index: 1340167 (C4 (Web Text))

Max Activating Token Index: 178

Text #19

Max Range: 4.4153. Min Range: -4.4153

Max Act: 2.1117. Min Act: -0.0000

Data Index: 301070 (C4 (Web Text))

Max Activating Token Index: 252