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

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

Neuron 519 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: 5.1597. Min Range: -5.1597

Max Act: 5.1597. Min Act: -0.0000

Data Index: 188132 (C4 (Web Text))

Max Activating Token Index: 51

Text #1

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.8485. Min Act: -0.0000

Data Index: 92055 (C4 (Web Text))

Max Activating Token Index: 471

Text #2

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.4282. Min Act: -0.0001

Data Index: 1021056 (C4 (Web Text))

Max Activating Token Index: 641

Text #3

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.5709. Min Act: -0.0000

Data Index: 870739 (C4 (Web Text))

Max Activating Token Index: 608

Text #4

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.1909. Min Act: -0.0001

Data Index: 328792 (C4 (Web Text))

Max Activating Token Index: 984

Text #5

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.0279. Min Act: -0.0000

Data Index: 610724 (C4 (Web Text))

Max Activating Token Index: 314

Text #6

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.1378. Min Act: -0.0000

Data Index: 174657 (C4 (Web Text))

Max Activating Token Index: 949

Text #7

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.1277. Min Act: -0.0001

Data Index: 426727 (C4 (Web Text))

Max Activating Token Index: 67

Text #8

Max Range: 5.1597. Min Range: -5.1597

Max Act: 4.0930. Min Act: -0.0001

Data Index: 358829 (C4 (Web Text))

Max Activating Token Index: 26

Text #9

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.6193. Min Act: -0.0000

Data Index: 1266415 (C4 (Web Text))

Max Activating Token Index: 64

Text #10

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.2352. Min Act: -0.0001

Data Index: 1537857 (Python Code)

Max Activating Token Index: 186

Text #11

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.4493. Min Act: -0.0001

Data Index: 1243996 (C4 (Web Text))

Max Activating Token Index: 733

Text #12

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.3910. Min Act: -0.0001

Data Index: 1219374 (C4 (Web Text))

Max Activating Token Index: 648

Text #13

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.1891. Min Act: -0.0001

Data Index: 1248693 (C4 (Web Text))

Max Activating Token Index: 166

Text #14

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.3897. Min Act: -0.0000

Data Index: 1161473 (C4 (Web Text))

Max Activating Token Index: 220

Text #15

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.1040. Min Act: -0.0001

Data Index: 1524781 (Python Code)

Max Activating Token Index: 576

Text #16

Max Range: 5.1597. Min Range: -5.1597

Max Act: 3.2231. Min Act: -0.0000

Data Index: 1149435 (C4 (Web Text))

Max Activating Token Index: 650

Text #17

Max Range: 5.1597. Min Range: -5.1597

Max Act: 2.9167. Min Act: -0.0000

Data Index: 509715 (C4 (Web Text))

Max Activating Token Index: 679

Text #18

Max Range: 5.1597. Min Range: -5.1597

Max Act: 2.6711. Min Act: -0.0000

Data Index: 767230 (C4 (Web Text))

Max Activating Token Index: 763

Text #19

Max Range: 5.1597. Min Range: -5.1597

Max Act: 2.9739. Min Act: -0.0001

Data Index: 1534050 (Python Code)

Max Activating Token Index: 503