Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1488 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 3.5982. Min Range: -3.5982

Max Act: 3.5982. Min Act: -0.0001

Data Index: 616387 (C4 (Web Text))

Max Activating Token Index: 848

Text #1

Max Range: 3.5982. Min Range: -3.5982

Max Act: 3.4818. Min Act: -0.0001

Data Index: 1336064 (C4 (Web Text))

Max Activating Token Index: 228

Text #2

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8952. Min Act: -0.0001

Data Index: 1516949 (Python Code)

Max Activating Token Index: 488

Text #3

Max Range: 3.5982. Min Range: -3.5982

Max Act: 3.0244. Min Act: -0.0001

Data Index: 375581 (C4 (Web Text))

Max Activating Token Index: 695

Text #4

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8428. Min Act: -0.0001

Data Index: 28557 (C4 (Web Text))

Max Activating Token Index: 694

Text #5

Max Range: 3.5982. Min Range: -3.5982

Max Act: 3.0744. Min Act: -0.0001

Data Index: 95605 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.9363. Min Act: -0.0001

Data Index: 985211 (C4 (Web Text))

Max Activating Token Index: 1001

Text #7

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8834. Min Act: -0.0001

Data Index: 874011 (C4 (Web Text))

Max Activating Token Index: 248

Text #8

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8862. Min Act: -0.0001

Data Index: 1604174 (Python Code)

Max Activating Token Index: 90

Text #9

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8558. Min Act: -0.0001

Data Index: 1558178 (Python Code)

Max Activating Token Index: 175

Text #10

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.7085. Min Act: -0.0001

Data Index: 1526446 (Python Code)

Max Activating Token Index: 586

Text #11

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.6944. Min Act: -0.0001

Data Index: 1534793 (Python Code)

Max Activating Token Index: 24

Text #12

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.9661. Min Act: -0.0001

Data Index: 956700 (C4 (Web Text))

Max Activating Token Index: 998

Text #13

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.8339. Min Act: -0.0001

Data Index: 181820 (C4 (Web Text))

Max Activating Token Index: 995

Text #14

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.7760. Min Act: -0.0001

Data Index: 390152 (C4 (Web Text))

Max Activating Token Index: 437

Text #15

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.7424. Min Act: -0.0001

Data Index: 636363 (C4 (Web Text))

Max Activating Token Index: 61

Text #16

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.7470. Min Act: -0.0001

Data Index: 490148 (C4 (Web Text))

Max Activating Token Index: 193

Text #17

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.6027. Min Act: -0.0001

Data Index: 662180 (C4 (Web Text))

Max Activating Token Index: 381

Text #18

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.6020. Min Act: -0.0001

Data Index: 185577 (C4 (Web Text))

Max Activating Token Index: 36

Text #19

Max Range: 3.5982. Min Range: -3.5982

Max Act: 2.7307. Min Act: -0.0001

Data Index: 1321037 (C4 (Web Text))

Max Activating Token Index: 240