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

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

Neuron 635 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.3405. Min Act: -0.0000

Data Index: 468546 (C4 (Web Text))

Max Activating Token Index: 937

Text #1

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.3489. Min Act: -0.0000

Data Index: 911586 (C4 (Web Text))

Max Activating Token Index: 641

Text #2

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.3174. Min Act: -0.0000

Data Index: 397899 (C4 (Web Text))

Max Activating Token Index: 567

Text #3

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.3146. Min Act: -0.0000

Data Index: 517930 (C4 (Web Text))

Max Activating Token Index: 977

Text #4

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2956. Min Act: -0.0000

Data Index: 1641966 (Python Code)

Max Activating Token Index: 810

Text #5

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2849. Min Act: -0.0000

Data Index: 616933 (C4 (Web Text))

Max Activating Token Index: 770

Text #6

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2990. Min Act: -0.0000

Data Index: 888485 (C4 (Web Text))

Max Activating Token Index: 216

Text #7

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2930. Min Act: -0.0000

Data Index: 1369009 (Python Code)

Max Activating Token Index: 600

Text #8

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2920. Min Act: -0.0000

Data Index: 1435622 (Python Code)

Max Activating Token Index: 150

Text #9

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2571. Min Act: -0.0000

Data Index: 733580 (C4 (Web Text))

Max Activating Token Index: 720

Text #10

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2643. Min Act: -0.0000

Data Index: 1452045 (Python Code)

Max Activating Token Index: 260

Text #11

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2845. Min Act: -0.0000

Data Index: 1277138 (C4 (Web Text))

Max Activating Token Index: 747

Text #12

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2544. Min Act: -0.0000

Data Index: 1583232 (Python Code)

Max Activating Token Index: 484

Text #13

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2541. Min Act: -0.0000

Data Index: 1559893 (Python Code)

Max Activating Token Index: 424

Text #14

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2598. Min Act: -0.0000

Data Index: 1466695 (Python Code)

Max Activating Token Index: 927

Text #15

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2565. Min Act: -0.0000

Data Index: 1285168 (C4 (Web Text))

Max Activating Token Index: 303

Text #16

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2485. Min Act: -0.0000

Data Index: 897402 (C4 (Web Text))

Max Activating Token Index: 839

Text #17

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2420. Min Act: -0.0000

Data Index: 1596023 (Python Code)

Max Activating Token Index: 36

Text #18

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2567. Min Act: -0.0000

Data Index: 199129 (C4 (Web Text))

Max Activating Token Index: 913

Text #19

Max Range: 0.3489. Min Range: -0.3489

Max Act: 0.2567. Min Act: -0.0000

Data Index: 392249 (C4 (Web Text))

Max Activating Token Index: 302