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

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

Neuron 4716 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.8989. Min Act: -0.0001

Data Index: 326222 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.5446. Min Act: -0.0001

Data Index: 610865 (C4 (Web Text))

Max Activating Token Index: 100

Text #2

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.2809. Min Act: -0.0001

Data Index: 379957 (C4 (Web Text))

Max Activating Token Index: 195

Text #3

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.1564. Min Act: -0.0001

Data Index: 971112 (C4 (Web Text))

Max Activating Token Index: 842

Text #4

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.3622. Min Act: -0.0001

Data Index: 684287 (C4 (Web Text))

Max Activating Token Index: 10

Text #5

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.3560. Min Act: -0.0001

Data Index: 319452 (C4 (Web Text))

Max Activating Token Index: 347

Text #6

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.4709. Min Act: -0.0001

Data Index: 589211 (C4 (Web Text))

Max Activating Token Index: 462

Text #7

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9080. Min Act: -0.0001

Data Index: 377357 (C4 (Web Text))

Max Activating Token Index: 157

Text #8

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.3762. Min Act: -0.0001

Data Index: 44778 (C4 (Web Text))

Max Activating Token Index: 895

Text #9

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.2367. Min Act: -0.0001

Data Index: 827615 (C4 (Web Text))

Max Activating Token Index: 442

Text #10

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.0661. Min Act: -0.0001

Data Index: 425121 (C4 (Web Text))

Max Activating Token Index: 601

Text #11

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9317. Min Act: -0.0001

Data Index: 20135 (C4 (Web Text))

Max Activating Token Index: 1018

Text #12

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.1559. Min Act: -0.0001

Data Index: 792061 (C4 (Web Text))

Max Activating Token Index: 841

Text #13

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.1704. Min Act: -0.0001

Data Index: 326412 (C4 (Web Text))

Max Activating Token Index: 39

Text #14

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9074. Min Act: -0.0001

Data Index: 575050 (C4 (Web Text))

Max Activating Token Index: 644

Text #15

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.1712. Min Act: -0.0001

Data Index: 290436 (C4 (Web Text))

Max Activating Token Index: 70

Text #16

Max Range: 3.8989. Min Range: -3.8989

Max Act: 3.4135. Min Act: -0.0001

Data Index: 911108 (C4 (Web Text))

Max Activating Token Index: 368

Text #17

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9498. Min Act: -0.0001

Data Index: 538365 (C4 (Web Text))

Max Activating Token Index: 519

Text #18

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9710. Min Act: -0.0001

Data Index: 1206650 (C4 (Web Text))

Max Activating Token Index: 361

Text #19

Max Range: 3.8989. Min Range: -3.8989

Max Act: 2.9779. Min Act: -0.0001

Data Index: 1036629 (C4 (Web Text))

Max Activating Token Index: 848