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

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

Neuron 1539 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0625. Min Act: -0.0000

Data Index: 405744 (C4 (Web Text))

Max Activating Token Index: 920

Text #1

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0575. Min Act: -0.0000

Data Index: 895400 (C4 (Web Text))

Max Activating Token Index: 430

Text #2

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0513. Min Act: -0.0001

Data Index: 87794 (C4 (Web Text))

Max Activating Token Index: 857

Text #3

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0527. Min Act: -0.0000

Data Index: 251677 (C4 (Web Text))

Max Activating Token Index: 922

Text #4

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0510. Min Act: -0.0000

Data Index: 1343361 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0534. Min Act: -0.0001

Data Index: 698851 (C4 (Web Text))

Max Activating Token Index: 782

Text #6

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0431. Min Act: -0.0001

Data Index: 131238 (C4 (Web Text))

Max Activating Token Index: 230

Text #7

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0466. Min Act: -0.0000

Data Index: 1206424 (C4 (Web Text))

Max Activating Token Index: 177

Text #8

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0426. Min Act: -0.0001

Data Index: 250526 (C4 (Web Text))

Max Activating Token Index: 156

Text #9

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0484. Min Act: -0.0001

Data Index: 53100 (C4 (Web Text))

Max Activating Token Index: 167

Text #10

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0463. Min Act: -0.0001

Data Index: 1102635 (C4 (Web Text))

Max Activating Token Index: 528

Text #11

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0474. Min Act: -0.0000

Data Index: 334882 (C4 (Web Text))

Max Activating Token Index: 1014

Text #12

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0437. Min Act: -0.0000

Data Index: 266392 (C4 (Web Text))

Max Activating Token Index: 312

Text #13

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0417. Min Act: -0.0001

Data Index: 1296220 (C4 (Web Text))

Max Activating Token Index: 748

Text #14

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0455. Min Act: -0.0001

Data Index: 532561 (C4 (Web Text))

Max Activating Token Index: 536

Text #15

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0399. Min Act: -0.0000

Data Index: 519808 (C4 (Web Text))

Max Activating Token Index: 146

Text #16

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0464. Min Act: -0.0001

Data Index: 472962 (C4 (Web Text))

Max Activating Token Index: 971

Text #17

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0419. Min Act: -0.0000

Data Index: 1346131 (C4 (Web Text))

Max Activating Token Index: 378

Text #18

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0423. Min Act: -0.0000

Data Index: 1530106 (Python Code)

Max Activating Token Index: 405

Text #19

Max Range: 0.0625. Min Range: -0.0625

Max Act: 0.0445. Min Act: -0.0001

Data Index: 1119894 (C4 (Web Text))

Max Activating Token Index: 276