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

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

Neuron 2630 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: 1.5584. Min Range: -1.5584

Max Act: 1.5441. Min Act: -0.0000

Data Index: 420213 (C4 (Web Text))

Max Activating Token Index: 161

Text #1

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.5221. Min Act: 0.0000

Data Index: 1276420 (C4 (Web Text))

Max Activating Token Index: 283

Text #2

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.5584. Min Act: -0.0000

Data Index: 799658 (C4 (Web Text))

Max Activating Token Index: 884

Text #3

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.3436. Min Act: -0.0000

Data Index: 1307265 (C4 (Web Text))

Max Activating Token Index: 463

Text #4

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.3344. Min Act: -0.0000

Data Index: 461610 (C4 (Web Text))

Max Activating Token Index: 548

Text #5

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.3147. Min Act: -0.0000

Data Index: 452097 (C4 (Web Text))

Max Activating Token Index: 998

Text #6

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.2374. Min Act: 0.0001

Data Index: 22554 (C4 (Web Text))

Max Activating Token Index: 408

Text #7

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.2124. Min Act: -0.0000

Data Index: 1603283 (Python Code)

Max Activating Token Index: 406

Text #8

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1965. Min Act: -0.0000

Data Index: 1399907 (Python Code)

Max Activating Token Index: 212

Text #9

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.2004. Min Act: 0.0000

Data Index: 130922 (C4 (Web Text))

Max Activating Token Index: 53

Text #10

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1995. Min Act: -0.0000

Data Index: 1443024 (Python Code)

Max Activating Token Index: 751

Text #11

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.0917. Min Act: -0.0000

Data Index: 1653557 (Python Code)

Max Activating Token Index: 947

Text #12

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1584. Min Act: -0.0000

Data Index: 797075 (C4 (Web Text))

Max Activating Token Index: 193

Text #13

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1527. Min Act: -0.0000

Data Index: 1170869 (C4 (Web Text))

Max Activating Token Index: 1020

Text #14

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1445. Min Act: -0.0000

Data Index: 308894 (C4 (Web Text))

Max Activating Token Index: 819

Text #15

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1527. Min Act: -0.0000

Data Index: 1653011 (Python Code)

Max Activating Token Index: 715

Text #16

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1378. Min Act: -0.0000

Data Index: 661774 (C4 (Web Text))

Max Activating Token Index: 87

Text #17

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1545. Min Act: -0.0000

Data Index: 1378278 (Python Code)

Max Activating Token Index: 904

Text #18

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.0763. Min Act: -0.0000

Data Index: 37057 (C4 (Web Text))

Max Activating Token Index: 485

Text #19

Max Range: 1.5584. Min Range: -1.5584

Max Act: 1.1324. Min Act: 0.0000

Data Index: 305600 (C4 (Web Text))

Max Activating Token Index: 166