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

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

Neuron 1934 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.9408. Min Act: -0.0000

Data Index: 251699 (C4 (Web Text))

Max Activating Token Index: 58

Text #1

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.8767. Min Act: -0.0000

Data Index: 1091956 (C4 (Web Text))

Max Activating Token Index: 985

Text #2

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.9304. Min Act: -0.0000

Data Index: 707441 (C4 (Web Text))

Max Activating Token Index: 240

Text #3

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.7506. Min Act: -0.0000

Data Index: 226355 (C4 (Web Text))

Max Activating Token Index: 297

Text #4

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.7084. Min Act: -0.0000

Data Index: 1565222 (Python Code)

Max Activating Token Index: 97

Text #5

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6197. Min Act: -0.0000

Data Index: 386585 (C4 (Web Text))

Max Activating Token Index: 199

Text #6

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6996. Min Act: -0.0000

Data Index: 1063238 (C4 (Web Text))

Max Activating Token Index: 287

Text #7

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.7421. Min Act: -0.0000

Data Index: 291171 (C4 (Web Text))

Max Activating Token Index: 44

Text #8

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6581. Min Act: -0.0000

Data Index: 461932 (C4 (Web Text))

Max Activating Token Index: 446

Text #9

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6361. Min Act: -0.0000

Data Index: 804370 (C4 (Web Text))

Max Activating Token Index: 182

Text #10

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6423. Min Act: -0.0000

Data Index: 775293 (C4 (Web Text))

Max Activating Token Index: 77

Text #11

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6635. Min Act: -0.0000

Data Index: 530834 (C4 (Web Text))

Max Activating Token Index: 410

Text #12

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6404. Min Act: -0.0000

Data Index: 1280651 (C4 (Web Text))

Max Activating Token Index: 975

Text #13

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6853. Min Act: -0.0000

Data Index: 522229 (C4 (Web Text))

Max Activating Token Index: 836

Text #14

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6468. Min Act: -0.0000

Data Index: 323236 (C4 (Web Text))

Max Activating Token Index: 835

Text #15

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6163. Min Act: -0.0000

Data Index: 604631 (C4 (Web Text))

Max Activating Token Index: 959

Text #16

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.5525. Min Act: -0.0000

Data Index: 983075 (C4 (Web Text))

Max Activating Token Index: 609

Text #17

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6658. Min Act: -0.0000

Data Index: 373588 (C4 (Web Text))

Max Activating Token Index: 147

Text #18

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6124. Min Act: -0.0000

Data Index: 1055775 (C4 (Web Text))

Max Activating Token Index: 850

Text #19

Max Range: 0.9408. Min Range: -0.9408

Max Act: 0.6729. Min Act: -0.0000

Data Index: 561531 (C4 (Web Text))

Max Activating Token Index: 773