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

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

Neuron 3631 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.5091. Min Act: -0.0001

Data Index: 853259 (C4 (Web Text))

Max Activating Token Index: 118

Text #1

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.5220. Min Act: -0.0001

Data Index: 1122807 (C4 (Web Text))

Max Activating Token Index: 387

Text #2

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4904. Min Act: -0.0001

Data Index: 597988 (C4 (Web Text))

Max Activating Token Index: 560

Text #3

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4594. Min Act: -0.0000

Data Index: 784918 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4790. Min Act: -0.0001

Data Index: 886725 (C4 (Web Text))

Max Activating Token Index: 791

Text #5

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4615. Min Act: -0.0000

Data Index: 1256150 (C4 (Web Text))

Max Activating Token Index: 610

Text #6

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4619. Min Act: -0.0000

Data Index: 312229 (C4 (Web Text))

Max Activating Token Index: 98

Text #7

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4261. Min Act: -0.0000

Data Index: 1213247 (C4 (Web Text))

Max Activating Token Index: 111

Text #8

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4365. Min Act: -0.0001

Data Index: 680701 (C4 (Web Text))

Max Activating Token Index: 409

Text #9

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4343. Min Act: -0.0000

Data Index: 470307 (C4 (Web Text))

Max Activating Token Index: 414

Text #10

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4469. Min Act: -0.0000

Data Index: 1017430 (C4 (Web Text))

Max Activating Token Index: 813

Text #11

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4558. Min Act: -0.0001

Data Index: 450399 (C4 (Web Text))

Max Activating Token Index: 190

Text #12

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4388. Min Act: -0.0001

Data Index: 1498891 (Python Code)

Max Activating Token Index: 446

Text #13

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4520. Min Act: -0.0001

Data Index: 346716 (C4 (Web Text))

Max Activating Token Index: 363

Text #14

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.3978. Min Act: -0.0001

Data Index: 1168237 (C4 (Web Text))

Max Activating Token Index: 848

Text #15

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4058. Min Act: -0.0001

Data Index: 1038720 (C4 (Web Text))

Max Activating Token Index: 948

Text #16

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.3938. Min Act: -0.0001

Data Index: 597908 (C4 (Web Text))

Max Activating Token Index: 646

Text #17

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.3863. Min Act: -0.0001

Data Index: 1619355 (Python Code)

Max Activating Token Index: 292

Text #18

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.4137. Min Act: -0.0001

Data Index: 1091483 (C4 (Web Text))

Max Activating Token Index: 640

Text #19

Max Range: 0.5220. Min Range: -0.5220

Max Act: 0.3850. Min Act: -0.0001

Data Index: 1286029 (C4 (Web Text))

Max Activating Token Index: 540