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

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

Neuron 3478 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2996. Min Act: -0.0000

Data Index: 924512 (C4 (Web Text))

Max Activating Token Index: 916

Text #1

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2871. Min Act: -0.0001

Data Index: 539316 (C4 (Web Text))

Max Activating Token Index: 870

Text #2

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2869. Min Act: -0.0001

Data Index: 1492324 (Python Code)

Max Activating Token Index: 298

Text #3

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2954. Min Act: -0.0000

Data Index: 436562 (C4 (Web Text))

Max Activating Token Index: 715

Text #4

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2726. Min Act: -0.0001

Data Index: 1134829 (C4 (Web Text))

Max Activating Token Index: 349

Text #5

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2284. Min Act: -0.0001

Data Index: 8465 (C4 (Web Text))

Max Activating Token Index: 405

Text #6

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2412. Min Act: -0.0000

Data Index: 945195 (C4 (Web Text))

Max Activating Token Index: 569

Text #7

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2262. Min Act: -0.0000

Data Index: 911878 (C4 (Web Text))

Max Activating Token Index: 522

Text #8

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2481. Min Act: -0.0000

Data Index: 1052660 (C4 (Web Text))

Max Activating Token Index: 778

Text #9

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2682. Min Act: -0.0000

Data Index: 7187 (C4 (Web Text))

Max Activating Token Index: 421

Text #10

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2602. Min Act: -0.0001

Data Index: 1563023 (Python Code)

Max Activating Token Index: 309

Text #11

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2508. Min Act: -0.0001

Data Index: 207483 (C4 (Web Text))

Max Activating Token Index: 718

Text #12

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2836. Min Act: -0.0000

Data Index: 1201129 (C4 (Web Text))

Max Activating Token Index: 298

Text #13

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2287. Min Act: -0.0001

Data Index: 1207804 (C4 (Web Text))

Max Activating Token Index: 382

Text #14

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2947. Min Act: -0.0001

Data Index: 895978 (C4 (Web Text))

Max Activating Token Index: 548

Text #15

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2286. Min Act: -0.0001

Data Index: 95750 (C4 (Web Text))

Max Activating Token Index: 487

Text #16

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2445. Min Act: -0.0001

Data Index: 1372942 (Python Code)

Max Activating Token Index: 670

Text #17

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2795. Min Act: -0.0000

Data Index: 14818 (C4 (Web Text))

Max Activating Token Index: 508

Text #18

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2692. Min Act: -0.0000

Data Index: 98215 (C4 (Web Text))

Max Activating Token Index: 898

Text #19

Max Range: 0.2996. Min Range: -0.2996

Max Act: 0.2803. Min Act: -0.0001

Data Index: 1350173 (C4 (Web Text))

Max Activating Token Index: 772