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

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

Neuron 4552 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.9521. Min Act: -0.0000

Data Index: 880433 (C4 (Web Text))

Max Activating Token Index: 102

Text #1

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.8395. Min Act: -0.0000

Data Index: 1086031 (C4 (Web Text))

Max Activating Token Index: 505

Text #2

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.8381. Min Act: -0.0000

Data Index: 842983 (C4 (Web Text))

Max Activating Token Index: 330

Text #3

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.8056. Min Act: -0.0000

Data Index: 392114 (C4 (Web Text))

Max Activating Token Index: 842

Text #4

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.7293. Min Act: -0.0000

Data Index: 1148400 (C4 (Web Text))

Max Activating Token Index: 125

Text #5

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.7697. Min Act: -0.0000

Data Index: 875182 (C4 (Web Text))

Max Activating Token Index: 486

Text #6

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.7411. Min Act: -0.0000

Data Index: 485727 (C4 (Web Text))

Max Activating Token Index: 944

Text #7

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6842. Min Act: -0.0000

Data Index: 248535 (C4 (Web Text))

Max Activating Token Index: 754

Text #8

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.7316. Min Act: -0.0000

Data Index: 732506 (C4 (Web Text))

Max Activating Token Index: 915

Text #9

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6777. Min Act: -0.0000

Data Index: 406681 (C4 (Web Text))

Max Activating Token Index: 173

Text #10

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6511. Min Act: -0.0000

Data Index: 1016583 (C4 (Web Text))

Max Activating Token Index: 767

Text #11

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.7085. Min Act: -0.0000

Data Index: 261698 (C4 (Web Text))

Max Activating Token Index: 608

Text #12

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6381. Min Act: -0.0000

Data Index: 276107 (C4 (Web Text))

Max Activating Token Index: 166

Text #13

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6608. Min Act: -0.0000

Data Index: 799089 (C4 (Web Text))

Max Activating Token Index: 593

Text #14

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6597. Min Act: 0.0001

Data Index: 1456167 (Python Code)

Max Activating Token Index: 119

Text #15

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6892. Min Act: 0.0000

Data Index: 241385 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6253. Min Act: -0.0000

Data Index: 1137168 (C4 (Web Text))

Max Activating Token Index: 274

Text #17

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6363. Min Act: -0.0000

Data Index: 1110039 (C4 (Web Text))

Max Activating Token Index: 876

Text #18

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.6326. Min Act: -0.0000

Data Index: 462056 (C4 (Web Text))

Max Activating Token Index: 643

Text #19

Max Range: 0.9521. Min Range: -0.9521

Max Act: 0.5890. Min Act: -0.0000

Data Index: 1035096 (C4 (Web Text))

Max Activating Token Index: 821