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

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

Neuron 5007 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.5967. Min Act: -0.0001

Data Index: 663783 (C4 (Web Text))

Max Activating Token Index: 315

Text #1

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4568. Min Act: -0.0001

Data Index: 209425 (C4 (Web Text))

Max Activating Token Index: 731

Text #2

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4680. Min Act: -0.0001

Data Index: 415257 (C4 (Web Text))

Max Activating Token Index: 351

Text #3

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4469. Min Act: -0.0001

Data Index: 1299182 (C4 (Web Text))

Max Activating Token Index: 745

Text #4

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4623. Min Act: -0.0001

Data Index: 129365 (C4 (Web Text))

Max Activating Token Index: 677

Text #5

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4589. Min Act: -0.0001

Data Index: 700777 (C4 (Web Text))

Max Activating Token Index: 493

Text #6

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4716. Min Act: -0.0001

Data Index: 1183600 (C4 (Web Text))

Max Activating Token Index: 268

Text #7

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4547. Min Act: -0.0001

Data Index: 248635 (C4 (Web Text))

Max Activating Token Index: 58

Text #8

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4547. Min Act: -0.0001

Data Index: 1098055 (C4 (Web Text))

Max Activating Token Index: 58

Text #9

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4475. Min Act: -0.0001

Data Index: 442043 (C4 (Web Text))

Max Activating Token Index: 709

Text #10

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4370. Min Act: -0.0001

Data Index: 999965 (C4 (Web Text))

Max Activating Token Index: 815

Text #11

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4791. Min Act: -0.0001

Data Index: 560 (C4 (Web Text))

Max Activating Token Index: 758

Text #12

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4418. Min Act: -0.0001

Data Index: 1043815 (C4 (Web Text))

Max Activating Token Index: 298

Text #13

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4321. Min Act: -0.0001

Data Index: 1140653 (C4 (Web Text))

Max Activating Token Index: 729

Text #14

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4331. Min Act: -0.0001

Data Index: 963275 (C4 (Web Text))

Max Activating Token Index: 407

Text #15

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4075. Min Act: -0.0001

Data Index: 1216663 (C4 (Web Text))

Max Activating Token Index: 114

Text #16

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.3972. Min Act: -0.0001

Data Index: 1392237 (Python Code)

Max Activating Token Index: 251

Text #17

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4316. Min Act: -0.0001

Data Index: 564758 (C4 (Web Text))

Max Activating Token Index: 502

Text #18

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4155. Min Act: -0.0001

Data Index: 788541 (C4 (Web Text))

Max Activating Token Index: 498

Text #19

Max Range: 0.5967. Min Range: -0.5967

Max Act: 0.4041. Min Act: -0.0001

Data Index: 39746 (C4 (Web Text))

Max Activating Token Index: 605