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

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

Neuron 3435 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: 9.8998. Min Range: -9.8998

Max Act: 9.8998. Min Act: -0.0001

Data Index: 647291 (C4 (Web Text))

Max Activating Token Index: 139

Text #1

Max Range: 9.8998. Min Range: -9.8998

Max Act: 8.0661. Min Act: -0.0001

Data Index: 1358814 (C4 (Web Text))

Max Activating Token Index: 737

Text #2

Max Range: 9.8998. Min Range: -9.8998

Max Act: 6.4670. Min Act: -0.0000

Data Index: 414183 (C4 (Web Text))

Max Activating Token Index: 908

Text #3

Max Range: 9.8998. Min Range: -9.8998

Max Act: 6.8055. Min Act: -0.0000

Data Index: 1208936 (C4 (Web Text))

Max Activating Token Index: 118

Text #4

Max Range: 9.8998. Min Range: -9.8998

Max Act: 5.8609. Min Act: -0.0000

Data Index: 211807 (C4 (Web Text))

Max Activating Token Index: 391

Text #5

Max Range: 9.8998. Min Range: -9.8998

Max Act: 5.7649. Min Act: -0.0001

Data Index: 833486 (C4 (Web Text))

Max Activating Token Index: 96

Text #6

Max Range: 9.8998. Min Range: -9.8998

Max Act: 5.2634. Min Act: -0.0001

Data Index: 653058 (C4 (Web Text))

Max Activating Token Index: 349

Text #7

Max Range: 9.8998. Min Range: -9.8998

Max Act: 4.2385. Min Act: -0.0000

Data Index: 341362 (C4 (Web Text))

Max Activating Token Index: 844

Text #8

Max Range: 9.8998. Min Range: -9.8998

Max Act: 3.7305. Min Act: -0.0000

Data Index: 613535 (C4 (Web Text))

Max Activating Token Index: 783

Text #9

Max Range: 9.8998. Min Range: -9.8998

Max Act: 4.8601. Min Act: 0.0000

Data Index: 716273 (C4 (Web Text))

Max Activating Token Index: 269

Text #10

Max Range: 9.8998. Min Range: -9.8998

Max Act: 4.2888. Min Act: -0.0000

Data Index: 440151 (C4 (Web Text))

Max Activating Token Index: 635

Text #11

Max Range: 9.8998. Min Range: -9.8998

Max Act: 4.7869. Min Act: -0.0001

Data Index: 1023923 (C4 (Web Text))

Max Activating Token Index: 360

Text #12

Max Range: 9.8998. Min Range: -9.8998

Max Act: 5.0356. Min Act: -0.0001

Data Index: 1037900 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 9.8998. Min Range: -9.8998

Max Act: 3.8663. Min Act: -0.0000

Data Index: 1332349 (C4 (Web Text))

Max Activating Token Index: 562

Text #14

Max Range: 9.8998. Min Range: -9.8998

Max Act: 5.1977. Min Act: -0.0001

Data Index: 1017599 (C4 (Web Text))

Max Activating Token Index: 608

Text #15

Max Range: 9.8998. Min Range: -9.8998

Max Act: 3.6174. Min Act: -0.0001

Data Index: 309659 (C4 (Web Text))

Max Activating Token Index: 1002

Text #16

Max Range: 9.8998. Min Range: -9.8998

Max Act: 2.6827. Min Act: 0.0001

Data Index: 1073502 (C4 (Web Text))

Max Activating Token Index: 263

Text #17

Max Range: 9.8998. Min Range: -9.8998

Max Act: 4.3592. Min Act: -0.0000

Data Index: 234991 (C4 (Web Text))

Max Activating Token Index: 829

Text #18

Max Range: 9.8998. Min Range: -9.8998

Max Act: 3.2668. Min Act: -0.0000

Data Index: 489063 (C4 (Web Text))

Max Activating Token Index: 44

Text #19

Max Range: 9.8998. Min Range: -9.8998

Max Act: 3.5403. Min Act: -0.0000

Data Index: 926315 (C4 (Web Text))

Max Activating Token Index: 371