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

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

Neuron 2724 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.5993. Min Act: -0.0000

Data Index: 389329 (C4 (Web Text))

Max Activating Token Index: 993

Text #1

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.4202. Min Act: -0.0000

Data Index: 613333 (C4 (Web Text))

Max Activating Token Index: 948

Text #2

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.6160. Min Act: -0.0000

Data Index: 797245 (C4 (Web Text))

Max Activating Token Index: 617

Text #3

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.2556. Min Act: -0.0000

Data Index: 451101 (C4 (Web Text))

Max Activating Token Index: 640

Text #4

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.1697. Min Act: -0.0000

Data Index: 1040229 (C4 (Web Text))

Max Activating Token Index: 94

Text #5

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.2927. Min Act: -0.0000

Data Index: 449777 (C4 (Web Text))

Max Activating Token Index: 716

Text #6

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.2668. Min Act: -0.0000

Data Index: 1426173 (Python Code)

Max Activating Token Index: 523

Text #7

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.1988. Min Act: -0.0000

Data Index: 546750 (C4 (Web Text))

Max Activating Token Index: 350

Text #8

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.2381. Min Act: -0.0000

Data Index: 814369 (C4 (Web Text))

Max Activating Token Index: 405

Text #9

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.1652. Min Act: -0.0000

Data Index: 192599 (C4 (Web Text))

Max Activating Token Index: 941

Text #10

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.1477. Min Act: -0.0000

Data Index: 296441 (C4 (Web Text))

Max Activating Token Index: 970

Text #11

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.1378. Min Act: -0.0000

Data Index: 1006925 (C4 (Web Text))

Max Activating Token Index: 434

Text #12

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0789. Min Act: -0.0000

Data Index: 1414615 (Python Code)

Max Activating Token Index: 604

Text #13

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0624. Min Act: -0.0000

Data Index: 1565844 (Python Code)

Max Activating Token Index: 795

Text #14

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0890. Min Act: -0.0000

Data Index: 1258353 (C4 (Web Text))

Max Activating Token Index: 608

Text #15

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0510. Min Act: -0.0000

Data Index: 734006 (C4 (Web Text))

Max Activating Token Index: 365

Text #16

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0102. Min Act: -0.0000

Data Index: 1129064 (C4 (Web Text))

Max Activating Token Index: 503

Text #17

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0299. Min Act: -0.0000

Data Index: 1087785 (C4 (Web Text))

Max Activating Token Index: 301

Text #18

Max Range: 1.6160. Min Range: -1.6160

Max Act: 0.9356. Min Act: -0.0000

Data Index: 860455 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 1.6160. Min Range: -1.6160

Max Act: 1.0012. Min Act: -0.0000

Data Index: 1550403 (Python Code)

Max Activating Token Index: 462