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

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

Neuron 2906 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0424. Min Act: -0.0001

Data Index: 437255 (C4 (Web Text))

Max Activating Token Index: 864

Text #1

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0375. Min Act: -0.0001

Data Index: 1128574 (C4 (Web Text))

Max Activating Token Index: 119

Text #2

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0348. Min Act: -0.0001

Data Index: 578052 (C4 (Web Text))

Max Activating Token Index: 684

Text #3

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0338. Min Act: -0.0002

Data Index: 467539 (C4 (Web Text))

Max Activating Token Index: 275

Text #4

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0334. Min Act: -0.0001

Data Index: 1376508 (Python Code)

Max Activating Token Index: 45

Text #5

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0323. Min Act: -0.0002

Data Index: 45801 (C4 (Web Text))

Max Activating Token Index: 895

Text #6

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0306. Min Act: -0.0001

Data Index: 1289114 (C4 (Web Text))

Max Activating Token Index: 781

Text #7

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0335. Min Act: -0.0001

Data Index: 1047208 (C4 (Web Text))

Max Activating Token Index: 801

Text #8

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0321. Min Act: -0.0001

Data Index: 1610256 (Python Code)

Max Activating Token Index: 29

Text #9

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0329. Min Act: -0.0001

Data Index: 209486 (C4 (Web Text))

Max Activating Token Index: 849

Text #10

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0306. Min Act: -0.0001

Data Index: 125527 (C4 (Web Text))

Max Activating Token Index: 353

Text #11

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0294. Min Act: -0.0001

Data Index: 229341 (C4 (Web Text))

Max Activating Token Index: 880

Text #12

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0342. Min Act: -0.0001

Data Index: 529679 (C4 (Web Text))

Max Activating Token Index: 166

Text #13

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0313. Min Act: -0.0002

Data Index: 127690 (C4 (Web Text))

Max Activating Token Index: 868

Text #14

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0293. Min Act: -0.0001

Data Index: 599070 (C4 (Web Text))

Max Activating Token Index: 596

Text #15

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0302. Min Act: -0.0001

Data Index: 1159990 (C4 (Web Text))

Max Activating Token Index: 876

Text #16

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0298. Min Act: -0.0002

Data Index: 258068 (C4 (Web Text))

Max Activating Token Index: 411

Text #17

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0291. Min Act: -0.0001

Data Index: 556774 (C4 (Web Text))

Max Activating Token Index: 942

Text #18

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0284. Min Act: -0.0001

Data Index: 1615340 (Python Code)

Max Activating Token Index: 1020

Text #19

Max Range: 0.0424. Min Range: -0.0424

Max Act: 0.0284. Min Act: -0.0002

Data Index: 963036 (C4 (Web Text))

Max Activating Token Index: 235