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

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

Neuron 3732 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1542. Min Act: -0.0000

Data Index: 890992 (C4 (Web Text))

Max Activating Token Index: 966

Text #1

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1288. Min Act: -0.0000

Data Index: 602022 (C4 (Web Text))

Max Activating Token Index: 438

Text #2

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1247. Min Act: -0.0000

Data Index: 681284 (C4 (Web Text))

Max Activating Token Index: 989

Text #3

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1309. Min Act: -0.0000

Data Index: 35219 (C4 (Web Text))

Max Activating Token Index: 348

Text #4

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1176. Min Act: -0.0000

Data Index: 604483 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1012. Min Act: -0.0000

Data Index: 1611316 (Python Code)

Max Activating Token Index: 262

Text #6

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1215. Min Act: -0.0000

Data Index: 1472078 (Python Code)

Max Activating Token Index: 842

Text #7

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1592. Min Act: -0.0001

Data Index: 672144 (C4 (Web Text))

Max Activating Token Index: 278

Text #8

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1112. Min Act: -0.0000

Data Index: 334751 (C4 (Web Text))

Max Activating Token Index: 403

Text #9

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1232. Min Act: -0.0000

Data Index: 1296791 (C4 (Web Text))

Max Activating Token Index: 946

Text #10

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1061. Min Act: -0.0000

Data Index: 570082 (C4 (Web Text))

Max Activating Token Index: 690

Text #11

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1095. Min Act: -0.0001

Data Index: 643800 (C4 (Web Text))

Max Activating Token Index: 715

Text #12

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1154. Min Act: -0.0000

Data Index: 1455042 (Python Code)

Max Activating Token Index: 971

Text #13

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1103. Min Act: -0.0000

Data Index: 540696 (C4 (Web Text))

Max Activating Token Index: 705

Text #14

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1098. Min Act: -0.0001

Data Index: 697995 (C4 (Web Text))

Max Activating Token Index: 266

Text #15

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1089. Min Act: -0.0000

Data Index: 1633002 (Python Code)

Max Activating Token Index: 527

Text #16

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1040. Min Act: -0.0000

Data Index: 75608 (C4 (Web Text))

Max Activating Token Index: 668

Text #17

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1102. Min Act: -0.0000

Data Index: 1528747 (Python Code)

Max Activating Token Index: 828

Text #18

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.0898. Min Act: -0.0001

Data Index: 1169722 (C4 (Web Text))

Max Activating Token Index: 864

Text #19

Max Range: 0.1592. Min Range: -0.1592

Max Act: 0.1014. Min Act: -0.0000

Data Index: 1001558 (C4 (Web Text))

Max Activating Token Index: 977