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

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

Neuron 2030 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.7239. Min Act: -0.0000

Data Index: 100543 (C4 (Web Text))

Max Activating Token Index: 727

Text #1

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.6324. Min Act: -0.0000

Data Index: 475551 (C4 (Web Text))

Max Activating Token Index: 549

Text #2

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.6670. Min Act: -0.0000

Data Index: 157350 (C4 (Web Text))

Max Activating Token Index: 612

Text #3

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5931. Min Act: -0.0000

Data Index: 200951 (C4 (Web Text))

Max Activating Token Index: 861

Text #4

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5656. Min Act: -0.0000

Data Index: 1120894 (C4 (Web Text))

Max Activating Token Index: 935

Text #5

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5264. Min Act: -0.0000

Data Index: 1282986 (C4 (Web Text))

Max Activating Token Index: 440

Text #6

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5996. Min Act: -0.0001

Data Index: 147854 (C4 (Web Text))

Max Activating Token Index: 240

Text #7

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.6016. Min Act: -0.0001

Data Index: 1564766 (Python Code)

Max Activating Token Index: 914

Text #8

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5486. Min Act: -0.0000

Data Index: 1271208 (C4 (Web Text))

Max Activating Token Index: 1022

Text #9

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4577. Min Act: -0.0001

Data Index: 271753 (C4 (Web Text))

Max Activating Token Index: 869

Text #10

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5172. Min Act: -0.0000

Data Index: 555192 (C4 (Web Text))

Max Activating Token Index: 744

Text #11

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4893. Min Act: -0.0001

Data Index: 909953 (C4 (Web Text))

Max Activating Token Index: 316

Text #12

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.5288. Min Act: -0.0000

Data Index: 1302556 (C4 (Web Text))

Max Activating Token Index: 640

Text #13

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4545. Min Act: -0.0000

Data Index: 89224 (C4 (Web Text))

Max Activating Token Index: 882

Text #14

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4952. Min Act: -0.0000

Data Index: 1068540 (C4 (Web Text))

Max Activating Token Index: 485

Text #15

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4853. Min Act: -0.0001

Data Index: 232834 (C4 (Web Text))

Max Activating Token Index: 630

Text #16

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4425. Min Act: -0.0001

Data Index: 1649867 (Python Code)

Max Activating Token Index: 661

Text #17

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4944. Min Act: -0.0000

Data Index: 767272 (C4 (Web Text))

Max Activating Token Index: 799

Text #18

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4667. Min Act: -0.0001

Data Index: 257596 (C4 (Web Text))

Max Activating Token Index: 883

Text #19

Max Range: 0.7239. Min Range: -0.7239

Max Act: 0.4612. Min Act: -0.0000

Data Index: 1164332 (C4 (Web Text))

Max Activating Token Index: 685