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

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

Neuron 4174 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.7445. Min Act: -0.0000

Data Index: 1468092 (Python Code)

Max Activating Token Index: 413

Text #1

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.6312. Min Act: -0.0000

Data Index: 80490 (C4 (Web Text))

Max Activating Token Index: 687

Text #2

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.5597. Min Act: -0.0000

Data Index: 1273929 (C4 (Web Text))

Max Activating Token Index: 197

Text #3

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.5249. Min Act: -0.0000

Data Index: 629719 (C4 (Web Text))

Max Activating Token Index: 524

Text #4

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.5395. Min Act: -0.0000

Data Index: 1428864 (Python Code)

Max Activating Token Index: 886

Text #5

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.5191. Min Act: -0.0000

Data Index: 1463185 (Python Code)

Max Activating Token Index: 794

Text #6

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4826. Min Act: -0.0000

Data Index: 925108 (C4 (Web Text))

Max Activating Token Index: 779

Text #7

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4899. Min Act: -0.0000

Data Index: 1412277 (Python Code)

Max Activating Token Index: 173

Text #8

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4845. Min Act: -0.0000

Data Index: 733484 (C4 (Web Text))

Max Activating Token Index: 678

Text #9

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4691. Min Act: -0.0000

Data Index: 1441986 (Python Code)

Max Activating Token Index: 219

Text #10

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4833. Min Act: -0.0000

Data Index: 1406268 (Python Code)

Max Activating Token Index: 561

Text #11

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4510. Min Act: -0.0001

Data Index: 1504730 (Python Code)

Max Activating Token Index: 846

Text #12

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4857. Min Act: -0.0000

Data Index: 1496331 (Python Code)

Max Activating Token Index: 477

Text #13

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4825. Min Act: -0.0000

Data Index: 942222 (C4 (Web Text))

Max Activating Token Index: 393

Text #14

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4790. Min Act: -0.0000

Data Index: 159802 (C4 (Web Text))

Max Activating Token Index: 266

Text #15

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4755. Min Act: -0.0000

Data Index: 787607 (C4 (Web Text))

Max Activating Token Index: 22

Text #16

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4433. Min Act: -0.0000

Data Index: 448963 (C4 (Web Text))

Max Activating Token Index: 506

Text #17

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4510. Min Act: -0.0000

Data Index: 820437 (C4 (Web Text))

Max Activating Token Index: 941

Text #18

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4605. Min Act: -0.0000

Data Index: 744014 (C4 (Web Text))

Max Activating Token Index: 790

Text #19

Max Range: 0.7445. Min Range: -0.7445

Max Act: 0.4506. Min Act: -0.0000

Data Index: 1297394 (C4 (Web Text))

Max Activating Token Index: 139