Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1082 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.1131. Min Act: -0.0001

Data Index: 990405 (C4 (Web Text))

Max Activating Token Index: 237

Text #1

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.1009. Min Act: -0.0002

Data Index: 815847 (C4 (Web Text))

Max Activating Token Index: 1022

Text #2

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.1008. Min Act: -0.0003

Data Index: 473273 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0963. Min Act: -0.0003

Data Index: 767534 (C4 (Web Text))

Max Activating Token Index: 824

Text #4

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0950. Min Act: -0.0002

Data Index: 325572 (C4 (Web Text))

Max Activating Token Index: 459

Text #5

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0960. Min Act: -0.0003

Data Index: 1202192 (C4 (Web Text))

Max Activating Token Index: 482

Text #6

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0902. Min Act: -0.0003

Data Index: 1140472 (C4 (Web Text))

Max Activating Token Index: 192

Text #7

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0898. Min Act: -0.0003

Data Index: 583435 (C4 (Web Text))

Max Activating Token Index: 47

Text #8

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0925. Min Act: -0.0002

Data Index: 1256320 (C4 (Web Text))

Max Activating Token Index: 368

Text #9

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0923. Min Act: -0.0002

Data Index: 765892 (C4 (Web Text))

Max Activating Token Index: 864

Text #10

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0876. Min Act: -0.0002

Data Index: 1310734 (C4 (Web Text))

Max Activating Token Index: 296

Text #11

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0891. Min Act: -0.0003

Data Index: 368521 (C4 (Web Text))

Max Activating Token Index: 21

Text #12

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0870. Min Act: -0.0003

Data Index: 518010 (C4 (Web Text))

Max Activating Token Index: 244

Text #13

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0872. Min Act: -0.0002

Data Index: 72710 (C4 (Web Text))

Max Activating Token Index: 353

Text #14

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0859. Min Act: -0.0003

Data Index: 1242306 (C4 (Web Text))

Max Activating Token Index: 83

Text #15

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0834. Min Act: -0.0003

Data Index: 61265 (C4 (Web Text))

Max Activating Token Index: 78

Text #16

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0887. Min Act: -0.0002

Data Index: 586618 (C4 (Web Text))

Max Activating Token Index: 477

Text #17

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0871. Min Act: -0.0003

Data Index: 485700 (C4 (Web Text))

Max Activating Token Index: 341

Text #18

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0828. Min Act: -0.0003

Data Index: 1001437 (C4 (Web Text))

Max Activating Token Index: 716

Text #19

Max Range: 0.1131. Min Range: -0.1131

Max Act: 0.0829. Min Act: -0.0002

Data Index: 98024 (C4 (Web Text))

Max Activating Token Index: 513