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

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

Neuron 3501 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.0850. Min Act: -0.0000

Data Index: 1141800 (C4 (Web Text))

Max Activating Token Index: 903

Text #1

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.0460. Min Act: 0.0001

Data Index: 424615 (C4 (Web Text))

Max Activating Token Index: 373

Text #2

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.2420. Min Act: -0.0000

Data Index: 727231 (C4 (Web Text))

Max Activating Token Index: 220

Text #3

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.0548. Min Act: -0.0000

Data Index: 1455012 (Python Code)

Max Activating Token Index: 507

Text #4

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.0624. Min Act: -0.0000

Data Index: 254268 (C4 (Web Text))

Max Activating Token Index: 228

Text #5

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.3980. Min Act: -0.0000

Data Index: 246250 (C4 (Web Text))

Max Activating Token Index: 216

Text #6

Max Range: 4.3980. Min Range: -4.3980

Max Act: 4.1229. Min Act: -0.0000

Data Index: 1034567 (C4 (Web Text))

Max Activating Token Index: 937

Text #7

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.5730. Min Act: -0.0000

Data Index: 900455 (C4 (Web Text))

Max Activating Token Index: 146

Text #8

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.6776. Min Act: -0.0000

Data Index: 1104206 (C4 (Web Text))

Max Activating Token Index: 949

Text #9

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.6606. Min Act: -0.0000

Data Index: 1064791 (C4 (Web Text))

Max Activating Token Index: 89

Text #10

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.8361. Min Act: -0.0000

Data Index: 243612 (C4 (Web Text))

Max Activating Token Index: 231

Text #11

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.6227. Min Act: -0.0000

Data Index: 481072 (C4 (Web Text))

Max Activating Token Index: 82

Text #12

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.7766. Min Act: -0.0000

Data Index: 1592004 (Python Code)

Max Activating Token Index: 441

Text #13

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.5838. Min Act: -0.0000

Data Index: 883927 (C4 (Web Text))

Max Activating Token Index: 388

Text #14

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.5406. Min Act: -0.0000

Data Index: 565036 (C4 (Web Text))

Max Activating Token Index: 959

Text #15

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.3316. Min Act: -0.0000

Data Index: 1365970 (Python Code)

Max Activating Token Index: 625

Text #16

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.3329. Min Act: -0.0000

Data Index: 720881 (C4 (Web Text))

Max Activating Token Index: 251

Text #17

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.0276. Min Act: -0.0000

Data Index: 758823 (C4 (Web Text))

Max Activating Token Index: 796

Text #18

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.5946. Min Act: -0.0000

Data Index: 1346393 (C4 (Web Text))

Max Activating Token Index: 283

Text #19

Max Range: 4.3980. Min Range: -4.3980

Max Act: 3.4436. Min Act: -0.0000

Data Index: 707825 (C4 (Web Text))

Max Activating Token Index: 293