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

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

Neuron 4786 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.8163. Min Act: -0.0000

Data Index: 370225 (C4 (Web Text))

Max Activating Token Index: 475

Text #1

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6948. Min Act: -0.0000

Data Index: 449720 (C4 (Web Text))

Max Activating Token Index: 817

Text #2

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.7045. Min Act: -0.0000

Data Index: 646130 (C4 (Web Text))

Max Activating Token Index: 167

Text #3

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6561. Min Act: -0.0000

Data Index: 638441 (C4 (Web Text))

Max Activating Token Index: 355

Text #4

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6499. Min Act: -0.0000

Data Index: 661879 (C4 (Web Text))

Max Activating Token Index: 237

Text #5

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6374. Min Act: -0.0000

Data Index: 555072 (C4 (Web Text))

Max Activating Token Index: 236

Text #6

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6155. Min Act: -0.0000

Data Index: 640740 (C4 (Web Text))

Max Activating Token Index: 237

Text #7

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5657. Min Act: -0.0000

Data Index: 466148 (C4 (Web Text))

Max Activating Token Index: 68

Text #8

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5756. Min Act: 0.0000

Data Index: 663000 (C4 (Web Text))

Max Activating Token Index: 121

Text #9

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5634. Min Act: 0.0001

Data Index: 317158 (C4 (Web Text))

Max Activating Token Index: 109

Text #10

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.6104. Min Act: -0.0000

Data Index: 749288 (C4 (Web Text))

Max Activating Token Index: 888

Text #11

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5379. Min Act: -0.0000

Data Index: 1224576 (C4 (Web Text))

Max Activating Token Index: 492

Text #12

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5635. Min Act: -0.0000

Data Index: 653730 (C4 (Web Text))

Max Activating Token Index: 602

Text #13

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5937. Min Act: 0.0000

Data Index: 984811 (C4 (Web Text))

Max Activating Token Index: 330

Text #14

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5111. Min Act: -0.0000

Data Index: 293043 (C4 (Web Text))

Max Activating Token Index: 306

Text #15

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5589. Min Act: -0.0000

Data Index: 308486 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5268. Min Act: -0.0000

Data Index: 16784 (C4 (Web Text))

Max Activating Token Index: 496

Text #17

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5206. Min Act: -0.0000

Data Index: 112995 (C4 (Web Text))

Max Activating Token Index: 829

Text #18

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.5304. Min Act: -0.0000

Data Index: 76347 (C4 (Web Text))

Max Activating Token Index: 868

Text #19

Max Range: 0.8163. Min Range: -0.8163

Max Act: 0.4979. Min Act: -0.0000

Data Index: 759443 (C4 (Web Text))

Max Activating Token Index: 978