Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2135 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0996. Min Act: -0.0001

Data Index: 1281055 (C4 (Web Text))

Max Activating Token Index: 196

Text #1

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0895. Min Act: -0.0001

Data Index: 878519 (C4 (Web Text))

Max Activating Token Index: 814

Text #2

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0919. Min Act: -0.0001

Data Index: 1145360 (C4 (Web Text))

Max Activating Token Index: 313

Text #3

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0961. Min Act: -0.0001

Data Index: 571283 (C4 (Web Text))

Max Activating Token Index: 434

Text #4

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0901. Min Act: -0.0001

Data Index: 324507 (C4 (Web Text))

Max Activating Token Index: 415

Text #5

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0860. Min Act: -0.0001

Data Index: 138868 (C4 (Web Text))

Max Activating Token Index: 121

Text #6

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0819. Min Act: -0.0001

Data Index: 1238648 (C4 (Web Text))

Max Activating Token Index: 368

Text #7

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0902. Min Act: -0.0001

Data Index: 832202 (C4 (Web Text))

Max Activating Token Index: 277

Text #8

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0745. Min Act: -0.0001

Data Index: 1533644 (Python Code)

Max Activating Token Index: 611

Text #9

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0846. Min Act: -0.0001

Data Index: 1174053 (C4 (Web Text))

Max Activating Token Index: 1008

Text #10

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0764. Min Act: -0.0001

Data Index: 1450706 (Python Code)

Max Activating Token Index: 789

Text #11

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0753. Min Act: -0.0001

Data Index: 414206 (C4 (Web Text))

Max Activating Token Index: 66

Text #12

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0746. Min Act: -0.0001

Data Index: 949350 (C4 (Web Text))

Max Activating Token Index: 511

Text #13

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0740. Min Act: -0.0001

Data Index: 428660 (C4 (Web Text))

Max Activating Token Index: 744

Text #14

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0713. Min Act: -0.0001

Data Index: 488474 (C4 (Web Text))

Max Activating Token Index: 202

Text #15

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0749. Min Act: -0.0001

Data Index: 1519771 (Python Code)

Max Activating Token Index: 798

Text #16

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0723. Min Act: -0.0001

Data Index: 51011 (C4 (Web Text))

Max Activating Token Index: 962

Text #17

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0717. Min Act: -0.0001

Data Index: 1321929 (C4 (Web Text))

Max Activating Token Index: 83

Text #18

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0695. Min Act: -0.0001

Data Index: 1346150 (C4 (Web Text))

Max Activating Token Index: 980

Text #19

Max Range: 0.0996. Min Range: -0.0996

Max Act: 0.0714. Min Act: -0.0001

Data Index: 846978 (C4 (Web Text))

Max Activating Token Index: 553