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

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

Neuron 1830 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2514. Min Range: -1.2514

Max Act: 1.2514. Min Act: -0.0000

Data Index: 768703 (C4 (Web Text))

Max Activating Token Index: 457

Text #1

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.8627. Min Act: -0.0000

Data Index: 1219733 (C4 (Web Text))

Max Activating Token Index: 21

Text #2

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.7589. Min Act: -0.0000

Data Index: 791751 (C4 (Web Text))

Max Activating Token Index: 108

Text #3

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.7733. Min Act: -0.0000

Data Index: 301446 (C4 (Web Text))

Max Activating Token Index: 665

Text #4

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.7226. Min Act: -0.0000

Data Index: 1326201 (C4 (Web Text))

Max Activating Token Index: 389

Text #5

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6865. Min Act: -0.0000

Data Index: 649180 (C4 (Web Text))

Max Activating Token Index: 711

Text #6

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6999. Min Act: -0.0000

Data Index: 718840 (C4 (Web Text))

Max Activating Token Index: 343

Text #7

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.7189. Min Act: -0.0000

Data Index: 621869 (C4 (Web Text))

Max Activating Token Index: 999

Text #8

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.7070. Min Act: -0.0000

Data Index: 834520 (C4 (Web Text))

Max Activating Token Index: 822

Text #9

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6870. Min Act: -0.0000

Data Index: 1218138 (C4 (Web Text))

Max Activating Token Index: 68

Text #10

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6838. Min Act: -0.0000

Data Index: 1466136 (Python Code)

Max Activating Token Index: 316

Text #11

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6538. Min Act: -0.0000

Data Index: 254816 (C4 (Web Text))

Max Activating Token Index: 740

Text #12

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6938. Min Act: -0.0000

Data Index: 1160560 (C4 (Web Text))

Max Activating Token Index: 830

Text #13

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6280. Min Act: -0.0000

Data Index: 1286350 (C4 (Web Text))

Max Activating Token Index: 545

Text #14

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6262. Min Act: -0.0000

Data Index: 1118651 (C4 (Web Text))

Max Activating Token Index: 537

Text #15

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.5857. Min Act: -0.0000

Data Index: 1575753 (Python Code)

Max Activating Token Index: 219

Text #16

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6416. Min Act: -0.0000

Data Index: 1149338 (C4 (Web Text))

Max Activating Token Index: 1021

Text #17

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6530. Min Act: -0.0000

Data Index: 185608 (C4 (Web Text))

Max Activating Token Index: 630

Text #18

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6094. Min Act: -0.0000

Data Index: 187855 (C4 (Web Text))

Max Activating Token Index: 776

Text #19

Max Range: 1.2514. Min Range: -1.2514

Max Act: 0.6216. Min Act: -0.0000

Data Index: 303692 (C4 (Web Text))

Max Activating Token Index: 574