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

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

Neuron 4821 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0252. Min Range: -1.0252

Max Act: 1.0252. Min Act: -0.0000

Data Index: 1090569 (C4 (Web Text))

Max Activating Token Index: 497

Text #1

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.9567. Min Act: -0.0000

Data Index: 1206978 (C4 (Web Text))

Max Activating Token Index: 215

Text #2

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.9270. Min Act: -0.0000

Data Index: 182522 (C4 (Web Text))

Max Activating Token Index: 610

Text #3

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.9016. Min Act: 0.0000

Data Index: 770428 (C4 (Web Text))

Max Activating Token Index: 703

Text #4

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.8634. Min Act: -0.0000

Data Index: 455447 (C4 (Web Text))

Max Activating Token Index: 203

Text #5

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.8523. Min Act: 0.0000

Data Index: 121504 (C4 (Web Text))

Max Activating Token Index: 108

Text #6

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.8381. Min Act: -0.0000

Data Index: 823025 (C4 (Web Text))

Max Activating Token Index: 464

Text #7

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7968. Min Act: -0.0000

Data Index: 784575 (C4 (Web Text))

Max Activating Token Index: 996

Text #8

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7874. Min Act: -0.0000

Data Index: 217379 (C4 (Web Text))

Max Activating Token Index: 395

Text #9

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7777. Min Act: -0.0000

Data Index: 726704 (C4 (Web Text))

Max Activating Token Index: 489

Text #10

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.8331. Min Act: 0.0000

Data Index: 901189 (C4 (Web Text))

Max Activating Token Index: 711

Text #11

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7220. Min Act: -0.0000

Data Index: 778756 (C4 (Web Text))

Max Activating Token Index: 751

Text #12

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7835. Min Act: -0.0000

Data Index: 1056697 (C4 (Web Text))

Max Activating Token Index: 146

Text #13

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7330. Min Act: -0.0000

Data Index: 1064007 (C4 (Web Text))

Max Activating Token Index: 677

Text #14

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7223. Min Act: -0.0000

Data Index: 1540571 (Python Code)

Max Activating Token Index: 172

Text #15

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7204. Min Act: -0.0000

Data Index: 1295849 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7378. Min Act: -0.0000

Data Index: 738792 (C4 (Web Text))

Max Activating Token Index: 368

Text #17

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7411. Min Act: -0.0000

Data Index: 450086 (C4 (Web Text))

Max Activating Token Index: 839

Text #18

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7164. Min Act: -0.0000

Data Index: 699226 (C4 (Web Text))

Max Activating Token Index: 341

Text #19

Max Range: 1.0252. Min Range: -1.0252

Max Act: 0.7516. Min Act: -0.0000

Data Index: 1101909 (C4 (Web Text))

Max Activating Token Index: 277