Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 562 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.9664. Min Act: -0.0001

Data Index: 1310348 (C4 (Web Text))

Max Activating Token Index: 600

Text #1

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.9241. Min Act: -0.0000

Data Index: 116674 (C4 (Web Text))

Max Activating Token Index: 308

Text #2

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.9312. Min Act: -0.0001

Data Index: 455385 (C4 (Web Text))

Max Activating Token Index: 579

Text #3

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.8530. Min Act: -0.0001

Data Index: 536384 (C4 (Web Text))

Max Activating Token Index: 160

Text #4

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.8257. Min Act: -0.0000

Data Index: 673514 (C4 (Web Text))

Max Activating Token Index: 454

Text #5

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.7463. Min Act: -0.0001

Data Index: 1022531 (C4 (Web Text))

Max Activating Token Index: 905

Text #6

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6828. Min Act: -0.0000

Data Index: 342867 (C4 (Web Text))

Max Activating Token Index: 544

Text #7

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6730. Min Act: -0.0000

Data Index: 1272867 (C4 (Web Text))

Max Activating Token Index: 1021

Text #8

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6970. Min Act: -0.0000

Data Index: 228038 (C4 (Web Text))

Max Activating Token Index: 761

Text #9

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6625. Min Act: -0.0001

Data Index: 811711 (C4 (Web Text))

Max Activating Token Index: 1009

Text #10

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6506. Min Act: -0.0000

Data Index: 141925 (C4 (Web Text))

Max Activating Token Index: 534

Text #11

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6352. Min Act: -0.0000

Data Index: 531046 (C4 (Web Text))

Max Activating Token Index: 635

Text #12

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6211. Min Act: -0.0000

Data Index: 419092 (C4 (Web Text))

Max Activating Token Index: 593

Text #13

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6299. Min Act: -0.0001

Data Index: 450843 (C4 (Web Text))

Max Activating Token Index: 273

Text #14

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6092. Min Act: -0.0001

Data Index: 585775 (C4 (Web Text))

Max Activating Token Index: 905

Text #15

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6237. Min Act: -0.0001

Data Index: 894810 (C4 (Web Text))

Max Activating Token Index: 289

Text #16

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.5925. Min Act: -0.0000

Data Index: 481639 (C4 (Web Text))

Max Activating Token Index: 1009

Text #17

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6138. Min Act: -0.0001

Data Index: 1342710 (C4 (Web Text))

Max Activating Token Index: 946

Text #18

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6043. Min Act: -0.0001

Data Index: 1080110 (C4 (Web Text))

Max Activating Token Index: 669

Text #19

Max Range: 0.9664. Min Range: -0.9664

Max Act: 0.6361. Min Act: -0.0000

Data Index: 442043 (C4 (Web Text))

Max Activating Token Index: 706