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

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

Neuron 4049 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.1576. Min Range: -7.1576

Max Act: 7.1576. Min Act: -0.0001

Data Index: 547054 (C4 (Web Text))

Max Activating Token Index: 482

Text #1

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.8616. Min Act: -0.0001

Data Index: 490570 (C4 (Web Text))

Max Activating Token Index: 936

Text #2

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.8185. Min Act: -0.0001

Data Index: 915935 (C4 (Web Text))

Max Activating Token Index: 596

Text #3

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.7474. Min Act: -0.0001

Data Index: 291986 (C4 (Web Text))

Max Activating Token Index: 50

Text #4

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.5379. Min Act: -0.0001

Data Index: 716529 (C4 (Web Text))

Max Activating Token Index: 571

Text #5

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.1640. Min Act: -0.0001

Data Index: 1179396 (C4 (Web Text))

Max Activating Token Index: 784

Text #6

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.2347. Min Act: -0.0001

Data Index: 787079 (C4 (Web Text))

Max Activating Token Index: 30

Text #7

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.4002. Min Act: -0.0001

Data Index: 1149704 (C4 (Web Text))

Max Activating Token Index: 396

Text #8

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.4128. Min Act: -0.0001

Data Index: 1193572 (C4 (Web Text))

Max Activating Token Index: 443

Text #9

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.7189. Min Act: -0.0001

Data Index: 1293119 (C4 (Web Text))

Max Activating Token Index: 581

Text #10

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.3323. Min Act: -0.0001

Data Index: 1183901 (C4 (Web Text))

Max Activating Token Index: 1022

Text #11

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.5960. Min Act: -0.0001

Data Index: 1287809 (C4 (Web Text))

Max Activating Token Index: 78

Text #12

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.8396. Min Act: -0.0001

Data Index: 1204500 (C4 (Web Text))

Max Activating Token Index: 81

Text #13

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.1834. Min Act: -0.0001

Data Index: 686261 (C4 (Web Text))

Max Activating Token Index: 32

Text #14

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.7215. Min Act: -0.0001

Data Index: 1229673 (C4 (Web Text))

Max Activating Token Index: 110

Text #15

Max Range: 7.1576. Min Range: -7.1576

Max Act: 6.1933. Min Act: -0.0001

Data Index: 593546 (C4 (Web Text))

Max Activating Token Index: 133

Text #16

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.6353. Min Act: -0.0001

Data Index: 1046672 (C4 (Web Text))

Max Activating Token Index: 569

Text #17

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.4275. Min Act: -0.0001

Data Index: 48578 (C4 (Web Text))

Max Activating Token Index: 855

Text #18

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.7870. Min Act: -0.0001

Data Index: 429531 (C4 (Web Text))

Max Activating Token Index: 437

Text #19

Max Range: 7.1576. Min Range: -7.1576

Max Act: 5.6118. Min Act: -0.0001

Data Index: 786918 (C4 (Web Text))

Max Activating Token Index: 357