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

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

Neuron 3592 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.9062. Min Act: -0.0000

Data Index: 896031 (C4 (Web Text))

Max Activating Token Index: 886

Text #1

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7608. Min Act: -0.0000

Data Index: 568611 (C4 (Web Text))

Max Activating Token Index: 229

Text #2

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7908. Min Act: -0.0000

Data Index: 876200 (C4 (Web Text))

Max Activating Token Index: 240

Text #3

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.8076. Min Act: -0.0000

Data Index: 998725 (C4 (Web Text))

Max Activating Token Index: 352

Text #4

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7733. Min Act: -0.0000

Data Index: 353956 (C4 (Web Text))

Max Activating Token Index: 356

Text #5

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7926. Min Act: -0.0000

Data Index: 140247 (C4 (Web Text))

Max Activating Token Index: 925

Text #6

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7586. Min Act: -0.0000

Data Index: 474423 (C4 (Web Text))

Max Activating Token Index: 559

Text #7

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7526. Min Act: -0.0000

Data Index: 225757 (C4 (Web Text))

Max Activating Token Index: 73

Text #8

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7654. Min Act: -0.0000

Data Index: 681435 (C4 (Web Text))

Max Activating Token Index: 630

Text #9

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7415. Min Act: 0.0001

Data Index: 57593 (C4 (Web Text))

Max Activating Token Index: 487

Text #10

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7246. Min Act: -0.0000

Data Index: 687320 (C4 (Web Text))

Max Activating Token Index: 38

Text #11

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7447. Min Act: 0.0000

Data Index: 1141784 (C4 (Web Text))

Max Activating Token Index: 274

Text #12

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7208. Min Act: -0.0000

Data Index: 1310889 (C4 (Web Text))

Max Activating Token Index: 596

Text #13

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7178. Min Act: 0.0003

Data Index: 510886 (C4 (Web Text))

Max Activating Token Index: 86

Text #14

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7322. Min Act: -0.0000

Data Index: 271565 (C4 (Web Text))

Max Activating Token Index: 778

Text #15

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7212. Min Act: -0.0000

Data Index: 984380 (C4 (Web Text))

Max Activating Token Index: 898

Text #16

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.6829. Min Act: -0.0000

Data Index: 1140864 (C4 (Web Text))

Max Activating Token Index: 861

Text #17

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.7064. Min Act: -0.0000

Data Index: 718338 (C4 (Web Text))

Max Activating Token Index: 901

Text #18

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.6782. Min Act: 0.0000

Data Index: 892131 (C4 (Web Text))

Max Activating Token Index: 669

Text #19

Max Range: 0.9062. Min Range: -0.9062

Max Act: 0.6674. Min Act: 0.0001

Data Index: 129095 (C4 (Web Text))

Max Activating Token Index: 976