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

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

Neuron 3934 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.4336. Min Act: -0.0001

Data Index: 969552 (C4 (Web Text))

Max Activating Token Index: 211

Text #1

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.3403. Min Act: -0.0000

Data Index: 784444 (C4 (Web Text))

Max Activating Token Index: 772

Text #2

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.3261. Min Act: -0.0000

Data Index: 673368 (C4 (Web Text))

Max Activating Token Index: 865

Text #3

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2991. Min Act: -0.0000

Data Index: 823664 (C4 (Web Text))

Max Activating Token Index: 62

Text #4

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2970. Min Act: -0.0001

Data Index: 1046865 (C4 (Web Text))

Max Activating Token Index: 248

Text #5

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2923. Min Act: -0.0000

Data Index: 893938 (C4 (Web Text))

Max Activating Token Index: 16

Text #6

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2888. Min Act: -0.0001

Data Index: 1320068 (C4 (Web Text))

Max Activating Token Index: 518

Text #7

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2941. Min Act: -0.0001

Data Index: 917337 (C4 (Web Text))

Max Activating Token Index: 973

Text #8

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2602. Min Act: -0.0001

Data Index: 914606 (C4 (Web Text))

Max Activating Token Index: 982

Text #9

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2657. Min Act: -0.0000

Data Index: 618309 (C4 (Web Text))

Max Activating Token Index: 487

Text #10

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2645. Min Act: -0.0001

Data Index: 413642 (C4 (Web Text))

Max Activating Token Index: 36

Text #11

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2717. Min Act: -0.0001

Data Index: 1493474 (Python Code)

Max Activating Token Index: 34

Text #12

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2675. Min Act: -0.0000

Data Index: 188791 (C4 (Web Text))

Max Activating Token Index: 27

Text #13

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2558. Min Act: -0.0001

Data Index: 107402 (C4 (Web Text))

Max Activating Token Index: 959

Text #14

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2585. Min Act: -0.0000

Data Index: 84179 (C4 (Web Text))

Max Activating Token Index: 87

Text #15

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2386. Min Act: -0.0001

Data Index: 1252406 (C4 (Web Text))

Max Activating Token Index: 569

Text #16

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2435. Min Act: -0.0001

Data Index: 1564682 (Python Code)

Max Activating Token Index: 44

Text #17

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2367. Min Act: -0.0000

Data Index: 986308 (C4 (Web Text))

Max Activating Token Index: 875

Text #18

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2439. Min Act: -0.0001

Data Index: 883894 (C4 (Web Text))

Max Activating Token Index: 395

Text #19

Max Range: 0.4336. Min Range: -0.4336

Max Act: 0.2231. Min Act: -0.0001

Data Index: 1000891 (C4 (Web Text))

Max Activating Token Index: 341