Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3421 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5772. Min Act: -0.0000

Data Index: 1576213 (Python Code)

Max Activating Token Index: 982

Text #1

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5564. Min Act: -0.0000

Data Index: 595238 (C4 (Web Text))

Max Activating Token Index: 754

Text #2

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5298. Min Act: -0.0000

Data Index: 1375472 (Python Code)

Max Activating Token Index: 159

Text #3

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5178. Min Act: -0.0000

Data Index: 1238542 (C4 (Web Text))

Max Activating Token Index: 314

Text #4

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4775. Min Act: -0.0000

Data Index: 130610 (C4 (Web Text))

Max Activating Token Index: 396

Text #5

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5241. Min Act: -0.0000

Data Index: 1230405 (C4 (Web Text))

Max Activating Token Index: 34

Text #6

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5032. Min Act: -0.0000

Data Index: 1577027 (Python Code)

Max Activating Token Index: 896

Text #7

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4962. Min Act: -0.0000

Data Index: 1014371 (C4 (Web Text))

Max Activating Token Index: 338

Text #8

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4854. Min Act: -0.0000

Data Index: 313918 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4453. Min Act: -0.0000

Data Index: 691011 (C4 (Web Text))

Max Activating Token Index: 266

Text #10

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.5087. Min Act: -0.0000

Data Index: 613875 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4676. Min Act: -0.0000

Data Index: 1541461 (Python Code)

Max Activating Token Index: 571

Text #12

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4823. Min Act: -0.0000

Data Index: 1306113 (C4 (Web Text))

Max Activating Token Index: 54

Text #13

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4412. Min Act: -0.0000

Data Index: 336260 (C4 (Web Text))

Max Activating Token Index: 183

Text #14

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4702. Min Act: -0.0000

Data Index: 546457 (C4 (Web Text))

Max Activating Token Index: 562

Text #15

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4501. Min Act: -0.0000

Data Index: 1463071 (Python Code)

Max Activating Token Index: 212

Text #16

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4573. Min Act: -0.0000

Data Index: 676023 (C4 (Web Text))

Max Activating Token Index: 311

Text #17

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4637. Min Act: -0.0000

Data Index: 1402087 (Python Code)

Max Activating Token Index: 801

Text #18

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4305. Min Act: -0.0000

Data Index: 280927 (C4 (Web Text))

Max Activating Token Index: 292

Text #19

Max Range: 0.5772. Min Range: -0.5772

Max Act: 0.4282. Min Act: -0.0000

Data Index: 1473839 (Python Code)

Max Activating Token Index: 889