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

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

Neuron 2808 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.5330. Min Act: -0.0001

Data Index: 635695 (C4 (Web Text))

Max Activating Token Index: 523

Text #1

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2879. Min Act: -0.0000

Data Index: 93559 (C4 (Web Text))

Max Activating Token Index: 942

Text #2

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2787. Min Act: 0.0000

Data Index: 333752 (C4 (Web Text))

Max Activating Token Index: 662

Text #3

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2784. Min Act: -0.0000

Data Index: 1000146 (C4 (Web Text))

Max Activating Token Index: 243

Text #4

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2601. Min Act: -0.0000

Data Index: 1581823 (Python Code)

Max Activating Token Index: 7

Text #5

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2497. Min Act: -0.0001

Data Index: 1407514 (Python Code)

Max Activating Token Index: 617

Text #6

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2787. Min Act: -0.0000

Data Index: 1129963 (C4 (Web Text))

Max Activating Token Index: 722

Text #7

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2321. Min Act: 0.0000

Data Index: 336689 (C4 (Web Text))

Max Activating Token Index: 8

Text #8

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2254. Min Act: -0.0001

Data Index: 1021569 (C4 (Web Text))

Max Activating Token Index: 714

Text #9

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2315. Min Act: -0.0001

Data Index: 1368040 (Python Code)

Max Activating Token Index: 973

Text #10

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2133. Min Act: -0.0000

Data Index: 200743 (C4 (Web Text))

Max Activating Token Index: 530

Text #11

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2135. Min Act: -0.0000

Data Index: 982759 (C4 (Web Text))

Max Activating Token Index: 605

Text #12

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2186. Min Act: -0.0000

Data Index: 452915 (C4 (Web Text))

Max Activating Token Index: 648

Text #13

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2145. Min Act: -0.0000

Data Index: 627215 (C4 (Web Text))

Max Activating Token Index: 27

Text #14

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2058. Min Act: -0.0000

Data Index: 1160279 (C4 (Web Text))

Max Activating Token Index: 430

Text #15

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.2105. Min Act: 0.0001

Data Index: 369267 (C4 (Web Text))

Max Activating Token Index: 438

Text #16

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.1891. Min Act: -0.0000

Data Index: 1348843 (C4 (Web Text))

Max Activating Token Index: 920

Text #17

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.1910. Min Act: -0.0001

Data Index: 1370331 (Python Code)

Max Activating Token Index: 350

Text #18

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.1858. Min Act: -0.0000

Data Index: 698029 (C4 (Web Text))

Max Activating Token Index: 773

Text #19

Max Range: 0.5330. Min Range: -0.5330

Max Act: 0.1788. Min Act: -0.0000

Data Index: 1228089 (C4 (Web Text))

Max Activating Token Index: 372