Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4292 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2846. Min Act: 0.0000

Data Index: 702809 (C4 (Web Text))

Max Activating Token Index: 791

Text #1

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2808. Min Act: 0.0000

Data Index: 349527 (C4 (Web Text))

Max Activating Token Index: 1009

Text #2

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2855. Min Act: -0.0000

Data Index: 215676 (C4 (Web Text))

Max Activating Token Index: 649

Text #3

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2513. Min Act: -0.0000

Data Index: 691065 (C4 (Web Text))

Max Activating Token Index: 272

Text #4

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2641. Min Act: 0.0000

Data Index: 23078 (C4 (Web Text))

Max Activating Token Index: 308

Text #5

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2418. Min Act: -0.0000

Data Index: 1048823 (C4 (Web Text))

Max Activating Token Index: 539

Text #6

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2569. Min Act: -0.0000

Data Index: 401538 (C4 (Web Text))

Max Activating Token Index: 385

Text #7

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2448. Min Act: -0.0000

Data Index: 206536 (C4 (Web Text))

Max Activating Token Index: 312

Text #8

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2449. Min Act: -0.0000

Data Index: 331463 (C4 (Web Text))

Max Activating Token Index: 354

Text #9

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2446. Min Act: -0.0000

Data Index: 1570040 (Python Code)

Max Activating Token Index: 955

Text #10

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2410. Min Act: 0.0000

Data Index: 1644282 (Python Code)

Max Activating Token Index: 448

Text #11

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2608. Min Act: -0.0000

Data Index: 1629012 (Python Code)

Max Activating Token Index: 112

Text #12

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2422. Min Act: -0.0000

Data Index: 945518 (C4 (Web Text))

Max Activating Token Index: 347

Text #13

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2257. Min Act: -0.0000

Data Index: 1375002 (Python Code)

Max Activating Token Index: 135

Text #14

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2103. Min Act: -0.0000

Data Index: 1204864 (C4 (Web Text))

Max Activating Token Index: 628

Text #15

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2202. Min Act: 0.0000

Data Index: 1456223 (Python Code)

Max Activating Token Index: 376

Text #16

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2194. Min Act: -0.0000

Data Index: 323485 (C4 (Web Text))

Max Activating Token Index: 435

Text #17

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2325. Min Act: -0.0000

Data Index: 1638432 (Python Code)

Max Activating Token Index: 54

Text #18

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2155. Min Act: 0.0000

Data Index: 1395471 (Python Code)

Max Activating Token Index: 304

Text #19

Max Range: 0.2855. Min Range: -0.2855

Max Act: 0.2162. Min Act: -0.0000

Data Index: 580390 (C4 (Web Text))

Max Activating Token Index: 81