Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 704 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.4539. Min Act: -0.0001

Data Index: 261349 (C4 (Web Text))

Max Activating Token Index: 734

Text #1

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3464. Min Act: -0.0001

Data Index: 573973 (C4 (Web Text))

Max Activating Token Index: 86

Text #2

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3188. Min Act: -0.0001

Data Index: 877045 (C4 (Web Text))

Max Activating Token Index: 918

Text #3

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3348. Min Act: -0.0000

Data Index: 178776 (C4 (Web Text))

Max Activating Token Index: 412

Text #4

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3689. Min Act: -0.0000

Data Index: 1512692 (Python Code)

Max Activating Token Index: 237

Text #5

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3001. Min Act: -0.0001

Data Index: 1507788 (Python Code)

Max Activating Token Index: 956

Text #6

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2948. Min Act: -0.0001

Data Index: 725415 (C4 (Web Text))

Max Activating Token Index: 976

Text #7

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2751. Min Act: -0.0000

Data Index: 1534336 (Python Code)

Max Activating Token Index: 448

Text #8

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2722. Min Act: -0.0001

Data Index: 24107 (C4 (Web Text))

Max Activating Token Index: 891

Text #9

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2989. Min Act: -0.0000

Data Index: 1467742 (Python Code)

Max Activating Token Index: 603

Text #10

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3078. Min Act: -0.0001

Data Index: 1091900 (C4 (Web Text))

Max Activating Token Index: 433

Text #11

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2943. Min Act: -0.0000

Data Index: 1375185 (Python Code)

Max Activating Token Index: 222

Text #12

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.3013. Min Act: -0.0001

Data Index: 1462028 (Python Code)

Max Activating Token Index: 857

Text #13

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2777. Min Act: -0.0000

Data Index: 1073809 (C4 (Web Text))

Max Activating Token Index: 771

Text #14

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2936. Min Act: -0.0001

Data Index: 623100 (C4 (Web Text))

Max Activating Token Index: 896

Text #15

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2872. Min Act: -0.0001

Data Index: 1414744 (Python Code)

Max Activating Token Index: 250

Text #16

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2843. Min Act: -0.0001

Data Index: 1586841 (Python Code)

Max Activating Token Index: 700

Text #17

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2717. Min Act: -0.0001

Data Index: 1446096 (Python Code)

Max Activating Token Index: 116

Text #18

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2783. Min Act: -0.0001

Data Index: 1530386 (Python Code)

Max Activating Token Index: 482

Text #19

Max Range: 0.4539. Min Range: -0.4539

Max Act: 0.2694. Min Act: -0.0001

Data Index: 1057870 (C4 (Web Text))

Max Activating Token Index: 998