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

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

Neuron 2809 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.5563. Min Act: -0.0000

Data Index: 854154 (C4 (Web Text))

Max Activating Token Index: 388

Text #1

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.5440. Min Act: -0.0000

Data Index: 1401724 (Python Code)

Max Activating Token Index: 32

Text #2

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.5347. Min Act: -0.0000

Data Index: 1419232 (Python Code)

Max Activating Token Index: 977

Text #3

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4936. Min Act: -0.0000

Data Index: 1642144 (Python Code)

Max Activating Token Index: 375

Text #4

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4573. Min Act: -0.0000

Data Index: 1594868 (Python Code)

Max Activating Token Index: 28

Text #5

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4853. Min Act: -0.0000

Data Index: 1479496 (Python Code)

Max Activating Token Index: 730

Text #6

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4462. Min Act: -0.0000

Data Index: 1455907 (Python Code)

Max Activating Token Index: 500

Text #7

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4362. Min Act: -0.0000

Data Index: 1210639 (C4 (Web Text))

Max Activating Token Index: 123

Text #8

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4636. Min Act: -0.0000

Data Index: 406853 (C4 (Web Text))

Max Activating Token Index: 155

Text #9

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4608. Min Act: -0.0000

Data Index: 1529646 (Python Code)

Max Activating Token Index: 193

Text #10

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4462. Min Act: -0.0000

Data Index: 873736 (C4 (Web Text))

Max Activating Token Index: 948

Text #11

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4416. Min Act: 0.0000

Data Index: 483656 (C4 (Web Text))

Max Activating Token Index: 22

Text #12

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4543. Min Act: -0.0000

Data Index: 14827 (C4 (Web Text))

Max Activating Token Index: 109

Text #13

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4371. Min Act: -0.0000

Data Index: 745980 (C4 (Web Text))

Max Activating Token Index: 228

Text #14

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4209. Min Act: -0.0000

Data Index: 350801 (C4 (Web Text))

Max Activating Token Index: 653

Text #15

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.3911. Min Act: -0.0000

Data Index: 92252 (C4 (Web Text))

Max Activating Token Index: 774

Text #16

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4432. Min Act: -0.0000

Data Index: 1642602 (Python Code)

Max Activating Token Index: 232

Text #17

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4257. Min Act: -0.0000

Data Index: 643488 (C4 (Web Text))

Max Activating Token Index: 802

Text #18

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4072. Min Act: -0.0000

Data Index: 74168 (C4 (Web Text))

Max Activating Token Index: 508

Text #19

Max Range: 0.5563. Min Range: -0.5563

Max Act: 0.4083. Min Act: -0.0000

Data Index: 1611715 (Python Code)

Max Activating Token Index: 1000