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

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

Neuron 3172 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.7980. Min Act: -0.0000

Data Index: 986409 (C4 (Web Text))

Max Activating Token Index: 320

Text #1

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.7614. Min Act: -0.0000

Data Index: 174789 (C4 (Web Text))

Max Activating Token Index: 99

Text #2

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.6977. Min Act: -0.0000

Data Index: 1308688 (C4 (Web Text))

Max Activating Token Index: 63

Text #3

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.5293. Min Act: -0.0000

Data Index: 102523 (C4 (Web Text))

Max Activating Token Index: 878

Text #4

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.6165. Min Act: -0.0000

Data Index: 734925 (C4 (Web Text))

Max Activating Token Index: 541

Text #5

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.6112. Min Act: -0.0000

Data Index: 197163 (C4 (Web Text))

Max Activating Token Index: 874

Text #6

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.4721. Min Act: -0.0000

Data Index: 410046 (C4 (Web Text))

Max Activating Token Index: 960

Text #7

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.5682. Min Act: -0.0000

Data Index: 1308207 (C4 (Web Text))

Max Activating Token Index: 826

Text #8

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.4189. Min Act: -0.0000

Data Index: 1319173 (C4 (Web Text))

Max Activating Token Index: 945

Text #9

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.4388. Min Act: -0.0000

Data Index: 1138867 (C4 (Web Text))

Max Activating Token Index: 760

Text #10

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.3768. Min Act: -0.0000

Data Index: 425433 (C4 (Web Text))

Max Activating Token Index: 960

Text #11

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.1522. Min Act: -0.0000

Data Index: 1357869 (C4 (Web Text))

Max Activating Token Index: 882

Text #12

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.2390. Min Act: -0.0000

Data Index: 4591 (C4 (Web Text))

Max Activating Token Index: 302

Text #13

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.3624. Min Act: -0.0000

Data Index: 316137 (C4 (Web Text))

Max Activating Token Index: 678

Text #14

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.0570. Min Act: -0.0000

Data Index: 71233 (C4 (Web Text))

Max Activating Token Index: 214

Text #15

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.1648. Min Act: -0.0000

Data Index: 786781 (C4 (Web Text))

Max Activating Token Index: 426

Text #16

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.1330. Min Act: -0.0000

Data Index: 203333 (C4 (Web Text))

Max Activating Token Index: 846

Text #17

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.0278. Min Act: -0.0000

Data Index: 822120 (C4 (Web Text))

Max Activating Token Index: 662

Text #18

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.2397. Min Act: -0.0000

Data Index: 587156 (C4 (Web Text))

Max Activating Token Index: 707

Text #19

Max Range: 2.7980. Min Range: -2.7980

Max Act: 2.2141. Min Act: -0.0000

Data Index: 508283 (C4 (Web Text))

Max Activating Token Index: 19