Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 70 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.5115. Min Act: -0.0001

Data Index: 683125 (C4 (Web Text))

Max Activating Token Index: 359

Text #1

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.5178. Min Act: -0.0001

Data Index: 1146818 (C4 (Web Text))

Max Activating Token Index: 558

Text #2

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4476. Min Act: -0.0001

Data Index: 1143695 (C4 (Web Text))

Max Activating Token Index: 796

Text #3

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4890. Min Act: -0.0001

Data Index: 450630 (C4 (Web Text))

Max Activating Token Index: 570

Text #4

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4971. Min Act: -0.0001

Data Index: 313380 (C4 (Web Text))

Max Activating Token Index: 683

Text #5

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4500. Min Act: -0.0001

Data Index: 1342169 (C4 (Web Text))

Max Activating Token Index: 349

Text #6

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4558. Min Act: -0.0001

Data Index: 435319 (C4 (Web Text))

Max Activating Token Index: 330

Text #7

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4271. Min Act: -0.0001

Data Index: 362934 (C4 (Web Text))

Max Activating Token Index: 149

Text #8

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4145. Min Act: -0.0001

Data Index: 1252653 (C4 (Web Text))

Max Activating Token Index: 861

Text #9

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4354. Min Act: -0.0001

Data Index: 1134877 (C4 (Web Text))

Max Activating Token Index: 732

Text #10

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4384. Min Act: -0.0001

Data Index: 454232 (C4 (Web Text))

Max Activating Token Index: 150

Text #11

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4501. Min Act: -0.0001

Data Index: 1345970 (C4 (Web Text))

Max Activating Token Index: 630

Text #12

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4212. Min Act: -0.0001

Data Index: 889221 (C4 (Web Text))

Max Activating Token Index: 141

Text #13

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4383. Min Act: -0.0001

Data Index: 662850 (C4 (Web Text))

Max Activating Token Index: 818

Text #14

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4322. Min Act: -0.0001

Data Index: 638243 (C4 (Web Text))

Max Activating Token Index: 540

Text #15

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4125. Min Act: -0.0001

Data Index: 850142 (C4 (Web Text))

Max Activating Token Index: 647

Text #16

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4032. Min Act: -0.0001

Data Index: 763155 (C4 (Web Text))

Max Activating Token Index: 319

Text #17

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4193. Min Act: -0.0001

Data Index: 657552 (C4 (Web Text))

Max Activating Token Index: 16

Text #18

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.4013. Min Act: -0.0001

Data Index: 1103424 (C4 (Web Text))

Max Activating Token Index: 957

Text #19

Max Range: 0.5178. Min Range: -0.5178

Max Act: 0.3794. Min Act: -0.0001

Data Index: 652265 (C4 (Web Text))

Max Activating Token Index: 738