Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 950 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.3199. Min Act: -0.0002

Data Index: 188562 (C4 (Web Text))

Max Activating Token Index: 277

Text #1

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.3208. Min Act: -0.0002

Data Index: 170264 (C4 (Web Text))

Max Activating Token Index: 916

Text #2

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.3092. Min Act: -0.0002

Data Index: 800593 (C4 (Web Text))

Max Activating Token Index: 372

Text #3

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.3115. Min Act: -0.0002

Data Index: 674516 (C4 (Web Text))

Max Activating Token Index: 765

Text #4

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2940. Min Act: -0.0002

Data Index: 92548 (C4 (Web Text))

Max Activating Token Index: 658

Text #5

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2948. Min Act: -0.0002

Data Index: 557793 (C4 (Web Text))

Max Activating Token Index: 717

Text #6

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2870. Min Act: -0.0002

Data Index: 121713 (C4 (Web Text))

Max Activating Token Index: 856

Text #7

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.3023. Min Act: -0.0002

Data Index: 847721 (C4 (Web Text))

Max Activating Token Index: 376

Text #8

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2877. Min Act: -0.0002

Data Index: 455384 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2636. Min Act: -0.0002

Data Index: 77970 (C4 (Web Text))

Max Activating Token Index: 228

Text #10

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2694. Min Act: -0.0002

Data Index: 836051 (C4 (Web Text))

Max Activating Token Index: 261

Text #11

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2733. Min Act: -0.0002

Data Index: 763831 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2907. Min Act: -0.0002

Data Index: 277770 (C4 (Web Text))

Max Activating Token Index: 467

Text #13

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2783. Min Act: -0.0002

Data Index: 96951 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2686. Min Act: -0.0002

Data Index: 564590 (C4 (Web Text))

Max Activating Token Index: 716

Text #15

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2606. Min Act: -0.0002

Data Index: 1214978 (C4 (Web Text))

Max Activating Token Index: 1018

Text #16

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2664. Min Act: -0.0002

Data Index: 836355 (C4 (Web Text))

Max Activating Token Index: 616

Text #17

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2565. Min Act: -0.0002

Data Index: 340552 (C4 (Web Text))

Max Activating Token Index: 667

Text #18

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2652. Min Act: -0.0002

Data Index: 324901 (C4 (Web Text))

Max Activating Token Index: 367

Text #19

Max Range: 0.3208. Min Range: -0.3208

Max Act: 0.2750. Min Act: -0.0002

Data Index: 1342982 (C4 (Web Text))

Max Activating Token Index: 822