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

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

Neuron 5330 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4945. Min Act: -0.0002

Data Index: 1487878 (Python Code)

Max Activating Token Index: 955

Text #1

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4923. Min Act: -0.0002

Data Index: 991202 (C4 (Web Text))

Max Activating Token Index: 616

Text #2

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.5239. Min Act: -0.0002

Data Index: 1457374 (Python Code)

Max Activating Token Index: 964

Text #3

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4732. Min Act: -0.0002

Data Index: 887919 (C4 (Web Text))

Max Activating Token Index: 861

Text #4

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4145. Min Act: -0.0002

Data Index: 1587754 (Python Code)

Max Activating Token Index: 801

Text #5

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4572. Min Act: -0.0002

Data Index: 807323 (C4 (Web Text))

Max Activating Token Index: 828

Text #6

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4118. Min Act: -0.0002

Data Index: 961001 (C4 (Web Text))

Max Activating Token Index: 827

Text #7

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.4031. Min Act: -0.0002

Data Index: 1434888 (Python Code)

Max Activating Token Index: 887

Text #8

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3821. Min Act: -0.0002

Data Index: 609377 (C4 (Web Text))

Max Activating Token Index: 696

Text #9

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3703. Min Act: -0.0002

Data Index: 1413407 (Python Code)

Max Activating Token Index: 862

Text #10

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3501. Min Act: -0.0002

Data Index: 1544932 (Python Code)

Max Activating Token Index: 718

Text #11

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3784. Min Act: -0.0002

Data Index: 645941 (C4 (Web Text))

Max Activating Token Index: 463

Text #12

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3614. Min Act: -0.0003

Data Index: 1245531 (C4 (Web Text))

Max Activating Token Index: 323

Text #13

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3502. Min Act: -0.0002

Data Index: 195005 (C4 (Web Text))

Max Activating Token Index: 869

Text #14

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3511. Min Act: -0.0002

Data Index: 1096455 (C4 (Web Text))

Max Activating Token Index: 236

Text #15

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3433. Min Act: -0.0002

Data Index: 487258 (C4 (Web Text))

Max Activating Token Index: 6

Text #16

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3075. Min Act: -0.0002

Data Index: 1351742 (C4 (Web Text))

Max Activating Token Index: 2

Text #17

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3075. Min Act: -0.0002

Data Index: 1168653 (C4 (Web Text))

Max Activating Token Index: 2

Text #18

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3075. Min Act: -0.0002

Data Index: 1200007 (C4 (Web Text))

Max Activating Token Index: 2

Text #19

Max Range: 1.5239. Min Range: -1.5239

Max Act: 1.3144. Min Act: -0.0002

Data Index: 1448537 (Python Code)

Max Activating Token Index: 982