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

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

Neuron 663 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4497. Min Range: -3.4497

Max Act: 3.4497. Min Act: -0.0001

Data Index: 677196 (C4 (Web Text))

Max Activating Token Index: 683

Text #1

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.5637. Min Act: -0.0001

Data Index: 86848 (C4 (Web Text))

Max Activating Token Index: 207

Text #2

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.5038. Min Act: -0.0001

Data Index: 1372022 (Python Code)

Max Activating Token Index: 330

Text #3

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.4356. Min Act: -0.0001

Data Index: 1513739 (Python Code)

Max Activating Token Index: 670

Text #4

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.3727. Min Act: -0.0001

Data Index: 771576 (C4 (Web Text))

Max Activating Token Index: 766

Text #5

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.2274. Min Act: -0.0001

Data Index: 1186742 (C4 (Web Text))

Max Activating Token Index: 32

Text #6

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.0087. Min Act: -0.0001

Data Index: 125340 (C4 (Web Text))

Max Activating Token Index: 124

Text #7

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.2131. Min Act: -0.0001

Data Index: 976632 (C4 (Web Text))

Max Activating Token Index: 698

Text #8

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.1319. Min Act: -0.0001

Data Index: 682994 (C4 (Web Text))

Max Activating Token Index: 677

Text #9

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.9691. Min Act: -0.0001

Data Index: 32834 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.0609. Min Act: -0.0001

Data Index: 676353 (C4 (Web Text))

Max Activating Token Index: 1021

Text #11

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.0076. Min Act: -0.0001

Data Index: 1010661 (C4 (Web Text))

Max Activating Token Index: 624

Text #12

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.1076. Min Act: -0.0001

Data Index: 199996 (C4 (Web Text))

Max Activating Token Index: 631

Text #13

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.9840. Min Act: -0.0001

Data Index: 1346275 (C4 (Web Text))

Max Activating Token Index: 939

Text #14

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.9517. Min Act: -0.0001

Data Index: 59358 (C4 (Web Text))

Max Activating Token Index: 696

Text #15

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.9339. Min Act: -0.0001

Data Index: 913167 (C4 (Web Text))

Max Activating Token Index: 281

Text #16

Max Range: 3.4497. Min Range: -3.4497

Max Act: 2.0047. Min Act: -0.0001

Data Index: 1150915 (C4 (Web Text))

Max Activating Token Index: 264

Text #17

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.8981. Min Act: -0.0001

Data Index: 756281 (C4 (Web Text))

Max Activating Token Index: 705

Text #18

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.9794. Min Act: -0.0001

Data Index: 1277242 (C4 (Web Text))

Max Activating Token Index: 1017

Text #19

Max Range: 3.4497. Min Range: -3.4497

Max Act: 1.8700. Min Act: -0.0001

Data Index: 1037237 (C4 (Web Text))

Max Activating Token Index: 1000