Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3394 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.6696. Min Act: -0.0000

Data Index: 1646088 (Python Code)

Max Activating Token Index: 823

Text #1

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.4589. Min Act: -0.0000

Data Index: 703457 (C4 (Web Text))

Max Activating Token Index: 251

Text #2

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.4043. Min Act: -0.0000

Data Index: 1439292 (Python Code)

Max Activating Token Index: 898

Text #3

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.4011. Min Act: -0.0000

Data Index: 1562677 (Python Code)

Max Activating Token Index: 997

Text #4

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.4024. Min Act: -0.0000

Data Index: 1288502 (C4 (Web Text))

Max Activating Token Index: 489

Text #5

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3606. Min Act: -0.0000

Data Index: 365382 (C4 (Web Text))

Max Activating Token Index: 113

Text #6

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3616. Min Act: -0.0000

Data Index: 1405535 (Python Code)

Max Activating Token Index: 32

Text #7

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3234. Min Act: -0.0000

Data Index: 1087738 (C4 (Web Text))

Max Activating Token Index: 790

Text #8

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3364. Min Act: -0.0000

Data Index: 1581173 (Python Code)

Max Activating Token Index: 697

Text #9

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3557. Min Act: -0.0000

Data Index: 380931 (C4 (Web Text))

Max Activating Token Index: 833

Text #10

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3006. Min Act: -0.0000

Data Index: 694241 (C4 (Web Text))

Max Activating Token Index: 199

Text #11

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3065. Min Act: -0.0000

Data Index: 1645415 (Python Code)

Max Activating Token Index: 53

Text #12

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2836. Min Act: -0.0000

Data Index: 1046877 (C4 (Web Text))

Max Activating Token Index: 167

Text #13

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.3011. Min Act: -0.0000

Data Index: 1272773 (C4 (Web Text))

Max Activating Token Index: 719

Text #14

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2918. Min Act: -0.0000

Data Index: 353373 (C4 (Web Text))

Max Activating Token Index: 91

Text #15

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2975. Min Act: -0.0000

Data Index: 424672 (C4 (Web Text))

Max Activating Token Index: 795

Text #16

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2886. Min Act: -0.0000

Data Index: 462780 (C4 (Web Text))

Max Activating Token Index: 114

Text #17

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2770. Min Act: -0.0000

Data Index: 1203113 (C4 (Web Text))

Max Activating Token Index: 40

Text #18

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2872. Min Act: -0.0000

Data Index: 16901 (C4 (Web Text))

Max Activating Token Index: 825

Text #19

Max Range: 0.6696. Min Range: -0.6696

Max Act: 0.2864. Min Act: -0.0000

Data Index: 234708 (C4 (Web Text))

Max Activating Token Index: 627