Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3812 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.9163. Min Act: -0.0001

Data Index: 531460 (C4 (Web Text))

Max Activating Token Index: 477

Text #1

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.8271. Min Act: -0.0001

Data Index: 667945 (C4 (Web Text))

Max Activating Token Index: 1005

Text #2

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.7744. Min Act: -0.0001

Data Index: 946507 (C4 (Web Text))

Max Activating Token Index: 1008

Text #3

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.7824. Min Act: -0.0001

Data Index: 688784 (C4 (Web Text))

Max Activating Token Index: 834

Text #4

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.7403. Min Act: -0.0001

Data Index: 28794 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.7063. Min Act: -0.0001

Data Index: 895231 (C4 (Web Text))

Max Activating Token Index: 768

Text #6

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6909. Min Act: -0.0001

Data Index: 1281634 (C4 (Web Text))

Max Activating Token Index: 18

Text #7

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.7076. Min Act: -0.0001

Data Index: 183633 (C4 (Web Text))

Max Activating Token Index: 759

Text #8

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6621. Min Act: -0.0001

Data Index: 1011570 (C4 (Web Text))

Max Activating Token Index: 602

Text #9

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6767. Min Act: -0.0001

Data Index: 1230793 (C4 (Web Text))

Max Activating Token Index: 953

Text #10

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6324. Min Act: -0.0000

Data Index: 802189 (C4 (Web Text))

Max Activating Token Index: 618

Text #11

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6341. Min Act: -0.0001

Data Index: 210782 (C4 (Web Text))

Max Activating Token Index: 986

Text #12

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6598. Min Act: -0.0001

Data Index: 150847 (C4 (Web Text))

Max Activating Token Index: 294

Text #13

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6243. Min Act: -0.0001

Data Index: 1393859 (Python Code)

Max Activating Token Index: 838

Text #14

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6280. Min Act: -0.0001

Data Index: 973128 (C4 (Web Text))

Max Activating Token Index: 943

Text #15

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6120. Min Act: -0.0000

Data Index: 424309 (C4 (Web Text))

Max Activating Token Index: 183

Text #16

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6315. Min Act: -0.0001

Data Index: 1472896 (Python Code)

Max Activating Token Index: 266

Text #17

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6466. Min Act: -0.0001

Data Index: 1030412 (C4 (Web Text))

Max Activating Token Index: 410

Text #18

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6213. Min Act: -0.0001

Data Index: 927225 (C4 (Web Text))

Max Activating Token Index: 280

Text #19

Max Range: 0.9163. Min Range: -0.9163

Max Act: 0.6233. Min Act: -0.0001

Data Index: 1336659 (C4 (Web Text))

Max Activating Token Index: 708