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

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

Neuron 1545 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.8208. Min Act: -0.0000

Data Index: 844949 (C4 (Web Text))

Max Activating Token Index: 987

Text #1

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.5553. Min Act: -0.0000

Data Index: 530808 (C4 (Web Text))

Max Activating Token Index: 166

Text #2

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.6651. Min Act: -0.0000

Data Index: 521822 (C4 (Web Text))

Max Activating Token Index: 56

Text #3

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.5865. Min Act: -0.0000

Data Index: 253008 (C4 (Web Text))

Max Activating Token Index: 420

Text #4

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.2340. Min Act: -0.0000

Data Index: 523468 (C4 (Web Text))

Max Activating Token Index: 990

Text #5

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.2234. Min Act: -0.0000

Data Index: 1416814 (Python Code)

Max Activating Token Index: 697

Text #6

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.2430. Min Act: -0.0000

Data Index: 1166779 (C4 (Web Text))

Max Activating Token Index: 736

Text #7

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.0315. Min Act: -0.0000

Data Index: 723398 (C4 (Web Text))

Max Activating Token Index: 13

Text #8

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.9895. Min Act: -0.0000

Data Index: 1054798 (C4 (Web Text))

Max Activating Token Index: 512

Text #9

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.0889. Min Act: -0.0000

Data Index: 954619 (C4 (Web Text))

Max Activating Token Index: 755

Text #10

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.8434. Min Act: -0.0000

Data Index: 679413 (C4 (Web Text))

Max Activating Token Index: 404

Text #11

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.0536. Min Act: -0.0000

Data Index: 469693 (C4 (Web Text))

Max Activating Token Index: 276

Text #12

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.8013. Min Act: -0.0000

Data Index: 1163314 (C4 (Web Text))

Max Activating Token Index: 773

Text #13

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.9241. Min Act: -0.0000

Data Index: 1401919 (Python Code)

Max Activating Token Index: 539

Text #14

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.0329. Min Act: -0.0000

Data Index: 697562 (C4 (Web Text))

Max Activating Token Index: 494

Text #15

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.9110. Min Act: -0.0000

Data Index: 123390 (C4 (Web Text))

Max Activating Token Index: 856

Text #16

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.9165. Min Act: -0.0000

Data Index: 807715 (C4 (Web Text))

Max Activating Token Index: 512

Text #17

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.9020. Min Act: -0.0000

Data Index: 1326793 (C4 (Web Text))

Max Activating Token Index: 459

Text #18

Max Range: 2.8208. Min Range: -2.8208

Max Act: 1.8886. Min Act: -0.0000

Data Index: 540634 (C4 (Web Text))

Max Activating Token Index: 953

Text #19

Max Range: 2.8208. Min Range: -2.8208

Max Act: 2.0080. Min Act: -0.0000

Data Index: 684647 (C4 (Web Text))

Max Activating Token Index: 835