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

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

Neuron 5086 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: 0.7496. Min Range: -0.7496

Max Act: 0.7496. Min Act: -0.0000

Data Index: 34153 (C4 (Web Text))

Max Activating Token Index: 897

Text #1

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.7450. Min Act: -0.0000

Data Index: 1267532 (C4 (Web Text))

Max Activating Token Index: 681

Text #2

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6654. Min Act: -0.0000

Data Index: 425992 (C4 (Web Text))

Max Activating Token Index: 878

Text #3

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5887. Min Act: -0.0000

Data Index: 1410434 (Python Code)

Max Activating Token Index: 981

Text #4

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5528. Min Act: -0.0000

Data Index: 1390565 (Python Code)

Max Activating Token Index: 623

Text #5

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.6117. Min Act: -0.0000

Data Index: 342059 (C4 (Web Text))

Max Activating Token Index: 915

Text #6

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5518. Min Act: -0.0000

Data Index: 1253083 (C4 (Web Text))

Max Activating Token Index: 312

Text #7

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5620. Min Act: -0.0000

Data Index: 628961 (C4 (Web Text))

Max Activating Token Index: 478

Text #8

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5597. Min Act: -0.0000

Data Index: 1127211 (C4 (Web Text))

Max Activating Token Index: 209

Text #9

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5258. Min Act: -0.0000

Data Index: 1248308 (C4 (Web Text))

Max Activating Token Index: 487

Text #10

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5179. Min Act: -0.0000

Data Index: 1521203 (Python Code)

Max Activating Token Index: 375

Text #11

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5187. Min Act: -0.0000

Data Index: 1534314 (Python Code)

Max Activating Token Index: 781

Text #12

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.4914. Min Act: -0.0001

Data Index: 252527 (C4 (Web Text))

Max Activating Token Index: 597

Text #13

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5054. Min Act: -0.0000

Data Index: 1497257 (Python Code)

Max Activating Token Index: 342

Text #14

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5347. Min Act: -0.0000

Data Index: 1347619 (C4 (Web Text))

Max Activating Token Index: 144

Text #15

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5183. Min Act: -0.0000

Data Index: 379561 (C4 (Web Text))

Max Activating Token Index: 925

Text #16

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.4805. Min Act: -0.0000

Data Index: 127405 (C4 (Web Text))

Max Activating Token Index: 988

Text #17

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5295. Min Act: -0.0001

Data Index: 362926 (C4 (Web Text))

Max Activating Token Index: 590

Text #18

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.5230. Min Act: -0.0000

Data Index: 1139756 (C4 (Web Text))

Max Activating Token Index: 155

Text #19

Max Range: 0.7496. Min Range: -0.7496

Max Act: 0.4492. Min Act: -0.0000

Data Index: 1075568 (C4 (Web Text))

Max Activating Token Index: 133