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

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

Neuron 5063 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3082. Min Range: -1.3082

Max Act: 1.3082. Min Act: -0.0000

Data Index: 895339 (C4 (Web Text))

Max Activating Token Index: 773

Text #1

Max Range: 1.3082. Min Range: -1.3082

Max Act: 1.2806. Min Act: -0.0000

Data Index: 492099 (C4 (Web Text))

Max Activating Token Index: 265

Text #2

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9044. Min Act: -0.0000

Data Index: 1324692 (C4 (Web Text))

Max Activating Token Index: 652

Text #3

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8500. Min Act: -0.0001

Data Index: 1117512 (C4 (Web Text))

Max Activating Token Index: 483

Text #4

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9112. Min Act: -0.0000

Data Index: 992979 (C4 (Web Text))

Max Activating Token Index: 905

Text #5

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9424. Min Act: -0.0000

Data Index: 1006848 (C4 (Web Text))

Max Activating Token Index: 887

Text #6

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9175. Min Act: -0.0001

Data Index: 1143593 (C4 (Web Text))

Max Activating Token Index: 589

Text #7

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9120. Min Act: -0.0000

Data Index: 1318161 (C4 (Web Text))

Max Activating Token Index: 642

Text #8

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8523. Min Act: -0.0000

Data Index: 273534 (C4 (Web Text))

Max Activating Token Index: 888

Text #9

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.9499. Min Act: -0.0000

Data Index: 440887 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8958. Min Act: -0.0001

Data Index: 1427855 (Python Code)

Max Activating Token Index: 499

Text #11

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8506. Min Act: -0.0001

Data Index: 1021857 (C4 (Web Text))

Max Activating Token Index: 182

Text #12

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8381. Min Act: -0.0000

Data Index: 166393 (C4 (Web Text))

Max Activating Token Index: 303

Text #13

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8855. Min Act: -0.0001

Data Index: 260923 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8047. Min Act: -0.0000

Data Index: 1588265 (Python Code)

Max Activating Token Index: 133

Text #15

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8742. Min Act: -0.0000

Data Index: 771040 (C4 (Web Text))

Max Activating Token Index: 828

Text #16

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.8113. Min Act: -0.0000

Data Index: 1409389 (Python Code)

Max Activating Token Index: 699

Text #17

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.7839. Min Act: -0.0001

Data Index: 1289030 (C4 (Web Text))

Max Activating Token Index: 80

Text #18

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.7801. Min Act: -0.0001

Data Index: 970479 (C4 (Web Text))

Max Activating Token Index: 171

Text #19

Max Range: 1.3082. Min Range: -1.3082

Max Act: 0.7288. Min Act: -0.0000

Data Index: 415400 (C4 (Web Text))

Max Activating Token Index: 282