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

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

Neuron 1589 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: 3.5747. Min Range: -3.5747

Max Act: 3.5747. Min Act: -0.0000

Data Index: 321788 (C4 (Web Text))

Max Activating Token Index: 722

Text #1

Max Range: 3.5747. Min Range: -3.5747

Max Act: 3.2543. Min Act: -0.0000

Data Index: 1307714 (C4 (Web Text))

Max Activating Token Index: 5

Text #2

Max Range: 3.5747. Min Range: -3.5747

Max Act: 2.8610. Min Act: -0.0000

Data Index: 513295 (C4 (Web Text))

Max Activating Token Index: 997

Text #3

Max Range: 3.5747. Min Range: -3.5747

Max Act: 2.5020. Min Act: -0.0000

Data Index: 713801 (C4 (Web Text))

Max Activating Token Index: 645

Text #4

Max Range: 3.5747. Min Range: -3.5747

Max Act: 2.3210. Min Act: -0.0000

Data Index: 1220177 (C4 (Web Text))

Max Activating Token Index: 922

Text #5

Max Range: 3.5747. Min Range: -3.5747

Max Act: 2.0347. Min Act: -0.0000

Data Index: 560158 (C4 (Web Text))

Max Activating Token Index: 68

Text #6

Max Range: 3.5747. Min Range: -3.5747

Max Act: 2.1413. Min Act: -0.0000

Data Index: 274565 (C4 (Web Text))

Max Activating Token Index: 873

Text #7

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.8723. Min Act: -0.0000

Data Index: 12676 (C4 (Web Text))

Max Activating Token Index: 350

Text #8

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.9246. Min Act: -0.0000

Data Index: 112867 (C4 (Web Text))

Max Activating Token Index: 133

Text #9

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.8560. Min Act: -0.0000

Data Index: 362259 (C4 (Web Text))

Max Activating Token Index: 510

Text #10

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.8243. Min Act: -0.0000

Data Index: 167530 (C4 (Web Text))

Max Activating Token Index: 949

Text #11

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.8010. Min Act: -0.0000

Data Index: 69817 (C4 (Web Text))

Max Activating Token Index: 724

Text #12

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.7629. Min Act: -0.0000

Data Index: 32397 (C4 (Web Text))

Max Activating Token Index: 288

Text #13

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.7308. Min Act: -0.0000

Data Index: 200252 (C4 (Web Text))

Max Activating Token Index: 820

Text #14

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.6022. Min Act: -0.0000

Data Index: 911632 (C4 (Web Text))

Max Activating Token Index: 368

Text #15

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.5902. Min Act: -0.0000

Data Index: 1046002 (C4 (Web Text))

Max Activating Token Index: 731

Text #16

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.7093. Min Act: -0.0000

Data Index: 1280581 (C4 (Web Text))

Max Activating Token Index: 727

Text #17

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.7104. Min Act: -0.0000

Data Index: 1393321 (Python Code)

Max Activating Token Index: 220

Text #18

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.6172. Min Act: -0.0000

Data Index: 377192 (C4 (Web Text))

Max Activating Token Index: 311

Text #19

Max Range: 3.5747. Min Range: -3.5747

Max Act: 1.5989. Min Act: -0.0000

Data Index: 245900 (C4 (Web Text))

Max Activating Token Index: 1022