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

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

Neuron 2514 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.3865. Min Range: -4.3865

Max Act: 4.3865. Min Act: -0.0000

Data Index: 556034 (C4 (Web Text))

Max Activating Token Index: 39

Text #1

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.8727. Min Act: -0.0000

Data Index: 423570 (C4 (Web Text))

Max Activating Token Index: 673

Text #2

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.7650. Min Act: -0.0000

Data Index: 1011296 (C4 (Web Text))

Max Activating Token Index: 879

Text #3

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.8606. Min Act: -0.0000

Data Index: 739769 (C4 (Web Text))

Max Activating Token Index: 975

Text #4

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.6070. Min Act: -0.0000

Data Index: 736007 (C4 (Web Text))

Max Activating Token Index: 707

Text #5

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.4887. Min Act: -0.0000

Data Index: 247947 (C4 (Web Text))

Max Activating Token Index: 472

Text #6

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.2401. Min Act: -0.0000

Data Index: 890077 (C4 (Web Text))

Max Activating Token Index: 340

Text #7

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.5965. Min Act: -0.0000

Data Index: 768830 (C4 (Web Text))

Max Activating Token Index: 87

Text #8

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.3527. Min Act: -0.0000

Data Index: 238323 (C4 (Web Text))

Max Activating Token Index: 730

Text #9

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.1572. Min Act: -0.0000

Data Index: 454901 (C4 (Web Text))

Max Activating Token Index: 429

Text #10

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0613. Min Act: -0.0000

Data Index: 1303502 (C4 (Web Text))

Max Activating Token Index: 369

Text #11

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.1464. Min Act: -0.0000

Data Index: 1589575 (Python Code)

Max Activating Token Index: 353

Text #12

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.1703. Min Act: -0.0000

Data Index: 76813 (C4 (Web Text))

Max Activating Token Index: 774

Text #13

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.1676. Min Act: -0.0000

Data Index: 1605178 (Python Code)

Max Activating Token Index: 320

Text #14

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0990. Min Act: -0.0000

Data Index: 510099 (C4 (Web Text))

Max Activating Token Index: 953

Text #15

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.2219. Min Act: -0.0000

Data Index: 567578 (C4 (Web Text))

Max Activating Token Index: 819

Text #16

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0879. Min Act: -0.0000

Data Index: 197081 (C4 (Web Text))

Max Activating Token Index: 659

Text #17

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0911. Min Act: -0.0000

Data Index: 601129 (C4 (Web Text))

Max Activating Token Index: 760

Text #18

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0851. Min Act: -0.0000

Data Index: 497596 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 4.3865. Min Range: -4.3865

Max Act: 2.0800. Min Act: -0.0000

Data Index: 1343743 (C4 (Web Text))

Max Activating Token Index: 1009