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

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

Neuron 755 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3854. Min Range: -5.3854

Max Act: 5.3854. Min Act: -0.0001

Data Index: 425828 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 5.3854. Min Range: -5.3854

Max Act: 5.3571. Min Act: -0.0001

Data Index: 1117146 (C4 (Web Text))

Max Activating Token Index: 288

Text #2

Max Range: 5.3854. Min Range: -5.3854

Max Act: 5.0778. Min Act: -0.0001

Data Index: 291950 (C4 (Web Text))

Max Activating Token Index: 314

Text #3

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.8676. Min Act: -0.0001

Data Index: 1241582 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.7568. Min Act: -0.0001

Data Index: 656762 (C4 (Web Text))

Max Activating Token Index: 32

Text #5

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.5798. Min Act: -0.0001

Data Index: 254972 (C4 (Web Text))

Max Activating Token Index: 519

Text #6

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.7051. Min Act: -0.0001

Data Index: 457729 (C4 (Web Text))

Max Activating Token Index: 1000

Text #7

Max Range: 5.3854. Min Range: -5.3854

Max Act: 5.0428. Min Act: -0.0001

Data Index: 600726 (C4 (Web Text))

Max Activating Token Index: 1022

Text #8

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.4161. Min Act: -0.0001

Data Index: 1214370 (C4 (Web Text))

Max Activating Token Index: 671

Text #9

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.4343. Min Act: -0.0001

Data Index: 829922 (C4 (Web Text))

Max Activating Token Index: 780

Text #10

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.1387. Min Act: -0.0001

Data Index: 1068089 (C4 (Web Text))

Max Activating Token Index: 114

Text #11

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.6243. Min Act: -0.0001

Data Index: 951217 (C4 (Web Text))

Max Activating Token Index: 370

Text #12

Max Range: 5.3854. Min Range: -5.3854

Max Act: 3.5498. Min Act: -0.0001

Data Index: 1052977 (C4 (Web Text))

Max Activating Token Index: 642

Text #13

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.2231. Min Act: -0.0001

Data Index: 381285 (C4 (Web Text))

Max Activating Token Index: 710

Text #14

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.0357. Min Act: -0.0001

Data Index: 1167228 (C4 (Web Text))

Max Activating Token Index: 596

Text #15

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.2201. Min Act: -0.0001

Data Index: 737611 (C4 (Web Text))

Max Activating Token Index: 674

Text #16

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.1775. Min Act: -0.0001

Data Index: 876288 (C4 (Web Text))

Max Activating Token Index: 78

Text #17

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.1863. Min Act: -0.0001

Data Index: 547021 (C4 (Web Text))

Max Activating Token Index: 245

Text #18

Max Range: 5.3854. Min Range: -5.3854

Max Act: 4.1267. Min Act: -0.0001

Data Index: 589211 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 5.3854. Min Range: -5.3854

Max Act: 3.9320. Min Act: -0.0001

Data Index: 1043173 (C4 (Web Text))

Max Activating Token Index: 403