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

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

Neuron 1898 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.9885. Min Act: -0.0001

Data Index: 1322664 (C4 (Web Text))

Max Activating Token Index: 436

Text #1

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.9026. Min Act: -0.0001

Data Index: 871367 (C4 (Web Text))

Max Activating Token Index: 750

Text #2

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.9126. Min Act: -0.0001

Data Index: 316465 (C4 (Web Text))

Max Activating Token Index: 730

Text #3

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.7310. Min Act: -0.0001

Data Index: 272118 (C4 (Web Text))

Max Activating Token Index: 763

Text #4

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.7452. Min Act: -0.0001

Data Index: 689086 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.7622. Min Act: -0.0001

Data Index: 216974 (C4 (Web Text))

Max Activating Token Index: 753

Text #6

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6785. Min Act: -0.0001

Data Index: 228544 (C4 (Web Text))

Max Activating Token Index: 631

Text #7

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.7039. Min Act: -0.0001

Data Index: 721815 (C4 (Web Text))

Max Activating Token Index: 888

Text #8

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6481. Min Act: -0.0001

Data Index: 1204065 (C4 (Web Text))

Max Activating Token Index: 286

Text #9

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6858. Min Act: -0.0001

Data Index: 123629 (C4 (Web Text))

Max Activating Token Index: 536

Text #10

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6335. Min Act: -0.0001

Data Index: 1058928 (C4 (Web Text))

Max Activating Token Index: 868

Text #11

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.5578. Min Act: -0.0001

Data Index: 32127 (C4 (Web Text))

Max Activating Token Index: 209

Text #12

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6356. Min Act: -0.0001

Data Index: 615689 (C4 (Web Text))

Max Activating Token Index: 239

Text #13

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6418. Min Act: -0.0001

Data Index: 1160353 (C4 (Web Text))

Max Activating Token Index: 757

Text #14

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6572. Min Act: -0.0001

Data Index: 1468642 (Python Code)

Max Activating Token Index: 239

Text #15

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.5948. Min Act: -0.0001

Data Index: 1505703 (Python Code)

Max Activating Token Index: 372

Text #16

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6459. Min Act: -0.0001

Data Index: 1261994 (C4 (Web Text))

Max Activating Token Index: 881

Text #17

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6277. Min Act: -0.0001

Data Index: 1561084 (Python Code)

Max Activating Token Index: 75

Text #18

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.5962. Min Act: -0.0001

Data Index: 300459 (C4 (Web Text))

Max Activating Token Index: 764

Text #19

Max Range: 0.9885. Min Range: -0.9885

Max Act: 0.6341. Min Act: -0.0001

Data Index: 414789 (C4 (Web Text))

Max Activating Token Index: 875