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

Dataset: The Pile

Neuron 3365 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.3756. Min Range: -9.3756

Max Act: 9.3756. Min Act: -0.0000

Data Index: 1586322 (The Pile)

Max Activating Token Index: 390

Text #1

Max Range: 9.3756. Min Range: -9.3756

Max Act: 9.0141. Min Act: -0.0000

Data Index: 1372807 (The Pile)

Max Activating Token Index: 265

Text #2

Max Range: 9.3756. Min Range: -9.3756

Max Act: 8.1157. Min Act: -0.0000

Data Index: 621232 (The Pile)

Max Activating Token Index: 690

Text #3

Max Range: 9.3756. Min Range: -9.3756

Max Act: 7.7896. Min Act: -0.0000

Data Index: 380906 (The Pile)

Max Activating Token Index: 440

Text #4

Max Range: 9.3756. Min Range: -9.3756

Max Act: 7.6487. Min Act: -0.0000

Data Index: 1354400 (The Pile)

Max Activating Token Index: 267

Text #5

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.6197. Min Act: -0.0000

Data Index: 833495 (The Pile)

Max Activating Token Index: 562

Text #6

Max Range: 9.3756. Min Range: -9.3756

Max Act: 7.6516. Min Act: -0.0000

Data Index: 549391 (The Pile)

Max Activating Token Index: 932

Text #7

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.6449. Min Act: -0.0000

Data Index: 421693 (The Pile)

Max Activating Token Index: 417

Text #8

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.9754. Min Act: -0.0000

Data Index: 1925166 (The Pile)

Max Activating Token Index: 490

Text #9

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.6335. Min Act: -0.0000

Data Index: 961872 (The Pile)

Max Activating Token Index: 849

Text #10

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.9708. Min Act: -0.0000

Data Index: 1343179 (The Pile)

Max Activating Token Index: 1006

Text #11

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.8066. Min Act: -0.0000

Data Index: 1425451 (The Pile)

Max Activating Token Index: 402

Text #12

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.5768. Min Act: -0.0000

Data Index: 822395 (The Pile)

Max Activating Token Index: 788

Text #13

Max Range: 9.3756. Min Range: -9.3756

Max Act: 7.0242. Min Act: -0.0000

Data Index: 537451 (The Pile)

Max Activating Token Index: 165

Text #14

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.5036. Min Act: -0.0000

Data Index: 1270241 (The Pile)

Max Activating Token Index: 948

Text #15

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.2008. Min Act: -0.0000

Data Index: 1390157 (The Pile)

Max Activating Token Index: 318

Text #16

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.8732. Min Act: -0.0000

Data Index: 871441 (The Pile)

Max Activating Token Index: 799

Text #17

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.2453. Min Act: -0.0000

Data Index: 1656992 (The Pile)

Max Activating Token Index: 707

Text #18

Max Range: 9.3756. Min Range: -9.3756

Max Act: 4.9159. Min Act: -0.0000

Data Index: 1226145 (The Pile)

Max Activating Token Index: 1000

Text #19

Max Range: 9.3756. Min Range: -9.3756

Max Act: 6.4079. Min Act: -0.0000

Data Index: 1186194 (The Pile)

Max Activating Token Index: 721