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

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

Neuron 770 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.7496. Min Act: -0.0000

Data Index: 218494 (C4 (Web Text))

Max Activating Token Index: 939

Text #1

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.6907. Min Act: -0.0000

Data Index: 358184 (C4 (Web Text))

Max Activating Token Index: 364

Text #2

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.6344. Min Act: -0.0000

Data Index: 1517560 (Python Code)

Max Activating Token Index: 713

Text #3

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.4401. Min Act: -0.0000

Data Index: 14790 (C4 (Web Text))

Max Activating Token Index: 942

Text #4

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.5001. Min Act: -0.0000

Data Index: 216373 (C4 (Web Text))

Max Activating Token Index: 877

Text #5

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.4434. Min Act: -0.0000

Data Index: 1137461 (C4 (Web Text))

Max Activating Token Index: 529

Text #6

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.4506. Min Act: -0.0000

Data Index: 882827 (C4 (Web Text))

Max Activating Token Index: 989

Text #7

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.3986. Min Act: -0.0000

Data Index: 817406 (C4 (Web Text))

Max Activating Token Index: 512

Text #8

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.2273. Min Act: -0.0000

Data Index: 775191 (C4 (Web Text))

Max Activating Token Index: 883

Text #9

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.2808. Min Act: -0.0000

Data Index: 308784 (C4 (Web Text))

Max Activating Token Index: 591

Text #10

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.1118. Min Act: -0.0000

Data Index: 325952 (C4 (Web Text))

Max Activating Token Index: 423

Text #11

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.0821. Min Act: -0.0000

Data Index: 556790 (C4 (Web Text))

Max Activating Token Index: 263

Text #12

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.1638. Min Act: -0.0000

Data Index: 267779 (C4 (Web Text))

Max Activating Token Index: 871

Text #13

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.1479. Min Act: -0.0000

Data Index: 1626591 (Python Code)

Max Activating Token Index: 259

Text #14

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.1268. Min Act: -0.0000

Data Index: 247473 (C4 (Web Text))

Max Activating Token Index: 147

Text #15

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.0028. Min Act: -0.0000

Data Index: 1642058 (Python Code)

Max Activating Token Index: 396

Text #16

Max Range: 2.7496. Min Range: -2.7496

Max Act: 1.9786. Min Act: -0.0000

Data Index: 991986 (C4 (Web Text))

Max Activating Token Index: 59

Text #17

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.0825. Min Act: -0.0000

Data Index: 1361382 (Python Code)

Max Activating Token Index: 795

Text #18

Max Range: 2.7496. Min Range: -2.7496

Max Act: 1.9512. Min Act: -0.0000

Data Index: 878733 (C4 (Web Text))

Max Activating Token Index: 276

Text #19

Max Range: 2.7496. Min Range: -2.7496

Max Act: 2.0648. Min Act: -0.0000

Data Index: 297945 (C4 (Web Text))

Max Activating Token Index: 545