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

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

Neuron 4066 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4794. Min Act: -0.0003

Data Index: 1016119 (C4 (Web Text))

Max Activating Token Index: 268

Text #1

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4328. Min Act: -0.0003

Data Index: 1457927 (Python Code)

Max Activating Token Index: 622

Text #2

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4946. Min Act: -0.0003

Data Index: 1475117 (Python Code)

Max Activating Token Index: 482

Text #3

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4821. Min Act: -0.0003

Data Index: 1474226 (Python Code)

Max Activating Token Index: 472

Text #4

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4660. Min Act: -0.0003

Data Index: 1391390 (Python Code)

Max Activating Token Index: 446

Text #5

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4643. Min Act: -0.0003

Data Index: 1611545 (Python Code)

Max Activating Token Index: 250

Text #6

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4579. Min Act: -0.0003

Data Index: 47535 (C4 (Web Text))

Max Activating Token Index: 887

Text #7

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.5177. Min Act: -0.0003

Data Index: 115813 (C4 (Web Text))

Max Activating Token Index: 508

Text #8

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4802. Min Act: -0.0003

Data Index: 342916 (C4 (Web Text))

Max Activating Token Index: 998

Text #9

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4762. Min Act: -0.0003

Data Index: 758515 (C4 (Web Text))

Max Activating Token Index: 3

Text #10

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.3734. Min Act: -0.0003

Data Index: 720135 (C4 (Web Text))

Max Activating Token Index: 463

Text #11

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4551. Min Act: -0.0003

Data Index: 1615023 (Python Code)

Max Activating Token Index: 336

Text #12

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4696. Min Act: -0.0003

Data Index: 584142 (C4 (Web Text))

Max Activating Token Index: 613

Text #13

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4366. Min Act: -0.0003

Data Index: 7999 (C4 (Web Text))

Max Activating Token Index: 23

Text #14

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4193. Min Act: -0.0003

Data Index: 82635 (C4 (Web Text))

Max Activating Token Index: 713

Text #15

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.3925. Min Act: -0.0003

Data Index: 1483152 (Python Code)

Max Activating Token Index: 733

Text #16

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.2693. Min Act: -0.0003

Data Index: 230303 (C4 (Web Text))

Max Activating Token Index: 539

Text #17

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.3919. Min Act: -0.0003

Data Index: 875894 (C4 (Web Text))

Max Activating Token Index: 470

Text #18

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.4158. Min Act: -0.0003

Data Index: 697548 (C4 (Web Text))

Max Activating Token Index: 887

Text #19

Max Range: 0.5177. Min Range: -0.5177

Max Act: 0.3041. Min Act: -0.0003

Data Index: 1284195 (C4 (Web Text))

Max Activating Token Index: 709