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

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

Neuron 4429 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: 3.6597. Min Range: -3.6597

Max Act: 3.6597. Min Act: -0.0000

Data Index: 1129116 (C4 (Web Text))

Max Activating Token Index: 833

Text #1

Max Range: 3.6597. Min Range: -3.6597

Max Act: 3.3931. Min Act: -0.0000

Data Index: 314912 (C4 (Web Text))

Max Activating Token Index: 752

Text #2

Max Range: 3.6597. Min Range: -3.6597

Max Act: 3.5024. Min Act: -0.0000

Data Index: 159356 (C4 (Web Text))

Max Activating Token Index: 644

Text #3

Max Range: 3.6597. Min Range: -3.6597

Max Act: 3.0460. Min Act: -0.0000

Data Index: 847448 (C4 (Web Text))

Max Activating Token Index: 360

Text #4

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.8738. Min Act: -0.0000

Data Index: 162297 (C4 (Web Text))

Max Activating Token Index: 120

Text #5

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.5306. Min Act: -0.0000

Data Index: 634068 (C4 (Web Text))

Max Activating Token Index: 746

Text #6

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.6461. Min Act: -0.0000

Data Index: 406538 (C4 (Web Text))

Max Activating Token Index: 88

Text #7

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.5938. Min Act: -0.0000

Data Index: 298984 (C4 (Web Text))

Max Activating Token Index: 923

Text #8

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.2704. Min Act: -0.0000

Data Index: 967098 (C4 (Web Text))

Max Activating Token Index: 208

Text #9

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.3704. Min Act: -0.0000

Data Index: 845318 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.2340. Min Act: -0.0000

Data Index: 797251 (C4 (Web Text))

Max Activating Token Index: 673

Text #11

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.1094. Min Act: -0.0000

Data Index: 1188840 (C4 (Web Text))

Max Activating Token Index: 305

Text #12

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.1929. Min Act: -0.0000

Data Index: 247598 (C4 (Web Text))

Max Activating Token Index: 966

Text #13

Max Range: 3.6597. Min Range: -3.6597

Max Act: 1.9448. Min Act: -0.0000

Data Index: 1173095 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.4256. Min Act: -0.0000

Data Index: 627193 (C4 (Web Text))

Max Activating Token Index: 716

Text #15

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.2652. Min Act: -0.0000

Data Index: 1466514 (Python Code)

Max Activating Token Index: 481

Text #16

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.0183. Min Act: -0.0000

Data Index: 1034936 (C4 (Web Text))

Max Activating Token Index: 864

Text #17

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.2054. Min Act: -0.0000

Data Index: 1410136 (Python Code)

Max Activating Token Index: 124

Text #18

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.0400. Min Act: -0.0000

Data Index: 1548914 (Python Code)

Max Activating Token Index: 244

Text #19

Max Range: 3.6597. Min Range: -3.6597

Max Act: 2.0697. Min Act: -0.0000

Data Index: 138571 (C4 (Web Text))

Max Activating Token Index: 111