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

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

Neuron 2445 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1711. Min Range: -1.1711

Max Act: 1.1711. Min Act: -0.0000

Data Index: 945140 (C4 (Web Text))

Max Activating Token Index: 456

Text #1

Max Range: 1.1711. Min Range: -1.1711

Max Act: 1.1355. Min Act: -0.0000

Data Index: 308852 (C4 (Web Text))

Max Activating Token Index: 554

Text #2

Max Range: 1.1711. Min Range: -1.1711

Max Act: 1.0523. Min Act: -0.0000

Data Index: 475018 (C4 (Web Text))

Max Activating Token Index: 915

Text #3

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.8722. Min Act: -0.0000

Data Index: 1575590 (Python Code)

Max Activating Token Index: 878

Text #4

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.9407. Min Act: -0.0000

Data Index: 1159227 (C4 (Web Text))

Max Activating Token Index: 194

Text #5

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.8323. Min Act: -0.0000

Data Index: 1319936 (C4 (Web Text))

Max Activating Token Index: 181

Text #6

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.8397. Min Act: -0.0000

Data Index: 253528 (C4 (Web Text))

Max Activating Token Index: 698

Text #7

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7708. Min Act: -0.0000

Data Index: 1479787 (Python Code)

Max Activating Token Index: 980

Text #8

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.8259. Min Act: -0.0000

Data Index: 856154 (C4 (Web Text))

Max Activating Token Index: 911

Text #9

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7671. Min Act: -0.0000

Data Index: 1082910 (C4 (Web Text))

Max Activating Token Index: 629

Text #10

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.8064. Min Act: 0.0000

Data Index: 177291 (C4 (Web Text))

Max Activating Token Index: 261

Text #11

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7803. Min Act: -0.0000

Data Index: 305650 (C4 (Web Text))

Max Activating Token Index: 55

Text #12

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7842. Min Act: -0.0000

Data Index: 1451639 (Python Code)

Max Activating Token Index: 876

Text #13

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7675. Min Act: -0.0000

Data Index: 249936 (C4 (Web Text))

Max Activating Token Index: 688

Text #14

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7830. Min Act: 0.0000

Data Index: 1313846 (C4 (Web Text))

Max Activating Token Index: 192

Text #15

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7215. Min Act: -0.0000

Data Index: 768920 (C4 (Web Text))

Max Activating Token Index: 825

Text #16

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7067. Min Act: -0.0000

Data Index: 1250268 (C4 (Web Text))

Max Activating Token Index: 967

Text #17

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7355. Min Act: -0.0000

Data Index: 683051 (C4 (Web Text))

Max Activating Token Index: 40

Text #18

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.7010. Min Act: -0.0000

Data Index: 1306776 (C4 (Web Text))

Max Activating Token Index: 196

Text #19

Max Range: 1.1711. Min Range: -1.1711

Max Act: 0.6772. Min Act: -0.0000

Data Index: 1596719 (Python Code)

Max Activating Token Index: 963