Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3365 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.8582. Min Act: -0.0000

Data Index: 1329765 (C4 (Web Text))

Max Activating Token Index: 933

Text #1

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6836. Min Act: -0.0000

Data Index: 336389 (C4 (Web Text))

Max Activating Token Index: 723

Text #2

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6201. Min Act: -0.0000

Data Index: 918825 (C4 (Web Text))

Max Activating Token Index: 761

Text #3

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6922. Min Act: -0.0000

Data Index: 1122214 (C4 (Web Text))

Max Activating Token Index: 87

Text #4

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.7109. Min Act: -0.0000

Data Index: 1315714 (C4 (Web Text))

Max Activating Token Index: 558

Text #5

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6806. Min Act: -0.0000

Data Index: 315268 (C4 (Web Text))

Max Activating Token Index: 33

Text #6

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6862. Min Act: -0.0000

Data Index: 678315 (C4 (Web Text))

Max Activating Token Index: 651

Text #7

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6582. Min Act: -0.0000

Data Index: 1428251 (Python Code)

Max Activating Token Index: 496

Text #8

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6451. Min Act: -0.0000

Data Index: 1322195 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6212. Min Act: -0.0000

Data Index: 399802 (C4 (Web Text))

Max Activating Token Index: 499

Text #10

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.5279. Min Act: -0.0000

Data Index: 201161 (C4 (Web Text))

Max Activating Token Index: 825

Text #11

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6351. Min Act: -0.0000

Data Index: 707165 (C4 (Web Text))

Max Activating Token Index: 548

Text #12

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6629. Min Act: -0.0000

Data Index: 1536956 (Python Code)

Max Activating Token Index: 998

Text #13

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6403. Min Act: -0.0000

Data Index: 1567123 (Python Code)

Max Activating Token Index: 370

Text #14

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.5929. Min Act: -0.0000

Data Index: 100689 (C4 (Web Text))

Max Activating Token Index: 289

Text #15

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.5731. Min Act: -0.0000

Data Index: 1386989 (Python Code)

Max Activating Token Index: 552

Text #16

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.5928. Min Act: -0.0000

Data Index: 1243580 (C4 (Web Text))

Max Activating Token Index: 165

Text #17

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6188. Min Act: -0.0000

Data Index: 1006828 (C4 (Web Text))

Max Activating Token Index: 166

Text #18

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6061. Min Act: -0.0000

Data Index: 1056827 (C4 (Web Text))

Max Activating Token Index: 379

Text #19

Max Range: 0.8582. Min Range: -0.8582

Max Act: 0.6424. Min Act: -0.0000

Data Index: 1069877 (C4 (Web Text))

Max Activating Token Index: 724