Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 520 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.9536. Min Act: -0.0000

Data Index: 1610180 (Python Code)

Max Activating Token Index: 318

Text #1

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.7449. Min Act: -0.0000

Data Index: 1595342 (Python Code)

Max Activating Token Index: 1019

Text #2

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.5146. Min Act: -0.0000

Data Index: 1048010 (C4 (Web Text))

Max Activating Token Index: 877

Text #3

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.3759. Min Act: -0.0000

Data Index: 1487495 (Python Code)

Max Activating Token Index: 990

Text #4

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.4669. Min Act: -0.0000

Data Index: 770045 (C4 (Web Text))

Max Activating Token Index: 954

Text #5

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.4196. Min Act: -0.0000

Data Index: 1402274 (Python Code)

Max Activating Token Index: 950

Text #6

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.3115. Min Act: 0.0000

Data Index: 1655756 (Python Code)

Max Activating Token Index: 558

Text #7

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.4299. Min Act: -0.0000

Data Index: 602799 (C4 (Web Text))

Max Activating Token Index: 61

Text #8

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.1131. Min Act: -0.0000

Data Index: 143494 (C4 (Web Text))

Max Activating Token Index: 399

Text #9

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.4587. Min Act: -0.0000

Data Index: 646719 (C4 (Web Text))

Max Activating Token Index: 948

Text #10

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.3065. Min Act: -0.0000

Data Index: 1557868 (Python Code)

Max Activating Token Index: 20

Text #11

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.3048. Min Act: -0.0000

Data Index: 1439397 (Python Code)

Max Activating Token Index: 716

Text #12

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.0604. Min Act: -0.0000

Data Index: 1402641 (Python Code)

Max Activating Token Index: 114

Text #13

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.2842. Min Act: -0.0000

Data Index: 1323419 (C4 (Web Text))

Max Activating Token Index: 457

Text #14

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.2626. Min Act: -0.0000

Data Index: 86529 (C4 (Web Text))

Max Activating Token Index: 690

Text #15

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.2086. Min Act: -0.0000

Data Index: 1554567 (Python Code)

Max Activating Token Index: 895

Text #16

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.0619. Min Act: 0.0000

Data Index: 540012 (C4 (Web Text))

Max Activating Token Index: 370

Text #17

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.3087. Min Act: -0.0000

Data Index: 478284 (C4 (Web Text))

Max Activating Token Index: 137

Text #18

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.1830. Min Act: -0.0000

Data Index: 564617 (C4 (Web Text))

Max Activating Token Index: 705

Text #19

Max Range: 2.9536. Min Range: -2.9536

Max Act: 2.1503. Min Act: -0.0000

Data Index: 1324376 (C4 (Web Text))

Max Activating Token Index: 867