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

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

Neuron 1732 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: 3.0880. Min Range: -3.0880

Max Act: 2.9049. Min Act: -0.0000

Data Index: 1410643 (Python Code)

Max Activating Token Index: 711

Text #1

Max Range: 3.0880. Min Range: -3.0880

Max Act: 3.0214. Min Act: -0.0000

Data Index: 252964 (C4 (Web Text))

Max Activating Token Index: 803

Text #2

Max Range: 3.0880. Min Range: -3.0880

Max Act: 3.0880. Min Act: -0.0000

Data Index: 804174 (C4 (Web Text))

Max Activating Token Index: 176

Text #3

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.8999. Min Act: -0.0000

Data Index: 906365 (C4 (Web Text))

Max Activating Token Index: 573

Text #4

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.8865. Min Act: -0.0000

Data Index: 844328 (C4 (Web Text))

Max Activating Token Index: 238

Text #5

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.7509. Min Act: -0.0000

Data Index: 1275747 (C4 (Web Text))

Max Activating Token Index: 456

Text #6

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.8777. Min Act: -0.0000

Data Index: 526515 (C4 (Web Text))

Max Activating Token Index: 183

Text #7

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.5871. Min Act: -0.0000

Data Index: 1559147 (Python Code)

Max Activating Token Index: 657

Text #8

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.6622. Min Act: -0.0000

Data Index: 1105408 (C4 (Web Text))

Max Activating Token Index: 333

Text #9

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.7127. Min Act: -0.0000

Data Index: 1526395 (Python Code)

Max Activating Token Index: 118

Text #10

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.6967. Min Act: -0.0000

Data Index: 1218800 (C4 (Web Text))

Max Activating Token Index: 293

Text #11

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.6314. Min Act: -0.0000

Data Index: 1575261 (Python Code)

Max Activating Token Index: 582

Text #12

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.4902. Min Act: -0.0000

Data Index: 1540400 (Python Code)

Max Activating Token Index: 561

Text #13

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.3765. Min Act: -0.0000

Data Index: 1623682 (Python Code)

Max Activating Token Index: 856

Text #14

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.4580. Min Act: -0.0000

Data Index: 417597 (C4 (Web Text))

Max Activating Token Index: 844

Text #15

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.2891. Min Act: -0.0000

Data Index: 1371026 (Python Code)

Max Activating Token Index: 164

Text #16

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.5277. Min Act: -0.0000

Data Index: 966916 (C4 (Web Text))

Max Activating Token Index: 877

Text #17

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.3842. Min Act: -0.0000

Data Index: 684492 (C4 (Web Text))

Max Activating Token Index: 771

Text #18

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.2414. Min Act: -0.0000

Data Index: 1267590 (C4 (Web Text))

Max Activating Token Index: 791

Text #19

Max Range: 3.0880. Min Range: -3.0880

Max Act: 2.3472. Min Act: -0.0000

Data Index: 1619697 (Python Code)

Max Activating Token Index: 198