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

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

Neuron 759 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0250. Min Range: -1.0250

Max Act: 1.0250. Min Act: -0.0001

Data Index: 73546 (C4 (Web Text))

Max Activating Token Index: 160

Text #1

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.9190. Min Act: -0.0001

Data Index: 989349 (C4 (Web Text))

Max Activating Token Index: 667

Text #2

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.9248. Min Act: -0.0001

Data Index: 1228685 (C4 (Web Text))

Max Activating Token Index: 587

Text #3

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.9129. Min Act: -0.0001

Data Index: 66799 (C4 (Web Text))

Max Activating Token Index: 671

Text #4

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.8300. Min Act: -0.0001

Data Index: 106589 (C4 (Web Text))

Max Activating Token Index: 614

Text #5

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.8681. Min Act: -0.0001

Data Index: 223111 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.8457. Min Act: -0.0001

Data Index: 351065 (C4 (Web Text))

Max Activating Token Index: 8

Text #7

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.8195. Min Act: -0.0001

Data Index: 1150734 (C4 (Web Text))

Max Activating Token Index: 672

Text #8

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7893. Min Act: -0.0001

Data Index: 5070 (C4 (Web Text))

Max Activating Token Index: 91

Text #9

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7515. Min Act: -0.0001

Data Index: 1205914 (C4 (Web Text))

Max Activating Token Index: 813

Text #10

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7957. Min Act: -0.0001

Data Index: 833646 (C4 (Web Text))

Max Activating Token Index: 600

Text #11

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7796. Min Act: -0.0001

Data Index: 1153606 (C4 (Web Text))

Max Activating Token Index: 191

Text #12

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7793. Min Act: -0.0001

Data Index: 1166480 (C4 (Web Text))

Max Activating Token Index: 802

Text #13

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7737. Min Act: -0.0001

Data Index: 931402 (C4 (Web Text))

Max Activating Token Index: 22

Text #14

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7707. Min Act: -0.0001

Data Index: 122083 (C4 (Web Text))

Max Activating Token Index: 993

Text #15

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7692. Min Act: -0.0001

Data Index: 245808 (C4 (Web Text))

Max Activating Token Index: 432

Text #16

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7622. Min Act: -0.0001

Data Index: 475477 (C4 (Web Text))

Max Activating Token Index: 797

Text #17

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7004. Min Act: -0.0001

Data Index: 1192084 (C4 (Web Text))

Max Activating Token Index: 928

Text #18

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.6996. Min Act: -0.0001

Data Index: 951319 (C4 (Web Text))

Max Activating Token Index: 253

Text #19

Max Range: 1.0250. Min Range: -1.0250

Max Act: 0.7479. Min Act: -0.0001

Data Index: 211379 (C4 (Web Text))

Max Activating Token Index: 264