Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1419 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.9908. Min Act: -0.0001

Data Index: 115121 (C4 (Web Text))

Max Activating Token Index: 82

Text #1

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.6425. Min Act: -0.0001

Data Index: 169988 (C4 (Web Text))

Max Activating Token Index: 747

Text #2

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.5628. Min Act: -0.0001

Data Index: 1399042 (Python Code)

Max Activating Token Index: 890

Text #3

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.6663. Min Act: -0.0001

Data Index: 492906 (C4 (Web Text))

Max Activating Token Index: 107

Text #4

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.5658. Min Act: -0.0001

Data Index: 872648 (C4 (Web Text))

Max Activating Token Index: 54

Text #5

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.4004. Min Act: -0.0001

Data Index: 408165 (C4 (Web Text))

Max Activating Token Index: 682

Text #6

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3430. Min Act: -0.0001

Data Index: 142572 (C4 (Web Text))

Max Activating Token Index: 529

Text #7

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3393. Min Act: -0.0001

Data Index: 203067 (C4 (Web Text))

Max Activating Token Index: 993

Text #8

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3677. Min Act: -0.0001

Data Index: 230533 (C4 (Web Text))

Max Activating Token Index: 128

Text #9

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.2735. Min Act: -0.0001

Data Index: 72819 (C4 (Web Text))

Max Activating Token Index: 942

Text #10

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3330. Min Act: -0.0001

Data Index: 526143 (C4 (Web Text))

Max Activating Token Index: 949

Text #11

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3718. Min Act: -0.0001

Data Index: 1598653 (Python Code)

Max Activating Token Index: 350

Text #12

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.2886. Min Act: -0.0001

Data Index: 1116119 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.2242. Min Act: -0.0001

Data Index: 795750 (C4 (Web Text))

Max Activating Token Index: 918

Text #14

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.2205. Min Act: -0.0001

Data Index: 434683 (C4 (Web Text))

Max Activating Token Index: 550

Text #15

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3448. Min Act: -0.0001

Data Index: 81265 (C4 (Web Text))

Max Activating Token Index: 828

Text #16

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.1599. Min Act: -0.0001

Data Index: 46639 (C4 (Web Text))

Max Activating Token Index: 353

Text #17

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3255. Min Act: -0.0001

Data Index: 1509042 (Python Code)

Max Activating Token Index: 505

Text #18

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.3066. Min Act: -0.0001

Data Index: 221334 (C4 (Web Text))

Max Activating Token Index: 837

Text #19

Max Range: 1.9908. Min Range: -1.9908

Max Act: 1.1570. Min Act: -0.0001

Data Index: 1338626 (C4 (Web Text))

Max Activating Token Index: 612