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

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

Neuron 2744 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.7609. Min Act: -0.0000

Data Index: 841350 (C4 (Web Text))

Max Activating Token Index: 28

Text #1

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.5924. Min Act: -0.0000

Data Index: 1289869 (C4 (Web Text))

Max Activating Token Index: 1008

Text #2

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.5558. Min Act: -0.0000

Data Index: 1499448 (Python Code)

Max Activating Token Index: 500

Text #3

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.5369. Min Act: -0.0000

Data Index: 1497957 (Python Code)

Max Activating Token Index: 840

Text #4

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.5042. Min Act: -0.0000

Data Index: 1431224 (Python Code)

Max Activating Token Index: 940

Text #5

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.4076. Min Act: -0.0000

Data Index: 120434 (C4 (Web Text))

Max Activating Token Index: 328

Text #6

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.3333. Min Act: -0.0000

Data Index: 737197 (C4 (Web Text))

Max Activating Token Index: 509

Text #7

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.3148. Min Act: -0.0000

Data Index: 1063984 (C4 (Web Text))

Max Activating Token Index: 65

Text #8

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.3402. Min Act: -0.0000

Data Index: 1587492 (Python Code)

Max Activating Token Index: 408

Text #9

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.3281. Min Act: -0.0000

Data Index: 1452877 (Python Code)

Max Activating Token Index: 718

Text #10

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.1924. Min Act: -0.0000

Data Index: 485021 (C4 (Web Text))

Max Activating Token Index: 279

Text #11

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.2436. Min Act: -0.0000

Data Index: 349956 (C4 (Web Text))

Max Activating Token Index: 407

Text #12

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.2406. Min Act: -0.0000

Data Index: 869181 (C4 (Web Text))

Max Activating Token Index: 67

Text #13

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.3478. Min Act: -0.0000

Data Index: 1649585 (Python Code)

Max Activating Token Index: 245

Text #14

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.1794. Min Act: -0.0000

Data Index: 911983 (C4 (Web Text))

Max Activating Token Index: 998

Text #15

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.2008. Min Act: -0.0000

Data Index: 24596 (C4 (Web Text))

Max Activating Token Index: 994

Text #16

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.2377. Min Act: -0.0000

Data Index: 1176365 (C4 (Web Text))

Max Activating Token Index: 635

Text #17

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.0907. Min Act: -0.0000

Data Index: 3420 (C4 (Web Text))

Max Activating Token Index: 367

Text #18

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.0357. Min Act: -0.0000

Data Index: 608141 (C4 (Web Text))

Max Activating Token Index: 680

Text #19

Max Range: 2.7609. Min Range: -2.7609

Max Act: 2.0301. Min Act: -0.0000

Data Index: 762904 (C4 (Web Text))

Max Activating Token Index: 670