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

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

Neuron 3472 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.4278. Min Act: -0.0000

Data Index: 1357545 (C4 (Web Text))

Max Activating Token Index: 61

Text #1

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.3853. Min Act: -0.0000

Data Index: 1003150 (C4 (Web Text))

Max Activating Token Index: 115

Text #2

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.3346. Min Act: -0.0000

Data Index: 334475 (C4 (Web Text))

Max Activating Token Index: 186

Text #3

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.3184. Min Act: -0.0000

Data Index: 195971 (C4 (Web Text))

Max Activating Token Index: 1018

Text #4

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.3167. Min Act: -0.0000

Data Index: 951178 (C4 (Web Text))

Max Activating Token Index: 442

Text #5

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2947. Min Act: -0.0000

Data Index: 590471 (C4 (Web Text))

Max Activating Token Index: 852

Text #6

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.3058. Min Act: -0.0000

Data Index: 909868 (C4 (Web Text))

Max Activating Token Index: 256

Text #7

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2948. Min Act: -0.0000

Data Index: 553397 (C4 (Web Text))

Max Activating Token Index: 995

Text #8

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2942. Min Act: -0.0000

Data Index: 422645 (C4 (Web Text))

Max Activating Token Index: 692

Text #9

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2780. Min Act: -0.0000

Data Index: 112645 (C4 (Web Text))

Max Activating Token Index: 578

Text #10

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2766. Min Act: -0.0000

Data Index: 456785 (C4 (Web Text))

Max Activating Token Index: 850

Text #11

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2666. Min Act: -0.0000

Data Index: 320191 (C4 (Web Text))

Max Activating Token Index: 921

Text #12

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2628. Min Act: -0.0000

Data Index: 1091625 (C4 (Web Text))

Max Activating Token Index: 689

Text #13

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2633. Min Act: -0.0000

Data Index: 1408264 (Python Code)

Max Activating Token Index: 236

Text #14

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2626. Min Act: -0.0000

Data Index: 751157 (C4 (Web Text))

Max Activating Token Index: 967

Text #15

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2596. Min Act: -0.0000

Data Index: 1110465 (C4 (Web Text))

Max Activating Token Index: 26

Text #16

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2596. Min Act: -0.0000

Data Index: 1536657 (Python Code)

Max Activating Token Index: 426

Text #17

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2668. Min Act: -0.0000

Data Index: 1325496 (C4 (Web Text))

Max Activating Token Index: 723

Text #18

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2570. Min Act: -0.0000

Data Index: 166480 (C4 (Web Text))

Max Activating Token Index: 525

Text #19

Max Range: 0.4278. Min Range: -0.4278

Max Act: 0.2510. Min Act: -0.0000

Data Index: 210606 (C4 (Web Text))

Max Activating Token Index: 589