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

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

Neuron 4396 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.3126. Min Act: -0.0000

Data Index: 527795 (C4 (Web Text))

Max Activating Token Index: 750

Text #1

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.3132. Min Act: -0.0000

Data Index: 506676 (C4 (Web Text))

Max Activating Token Index: 65

Text #2

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2820. Min Act: -0.0000

Data Index: 127556 (C4 (Web Text))

Max Activating Token Index: 311

Text #3

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2585. Min Act: -0.0000

Data Index: 366178 (C4 (Web Text))

Max Activating Token Index: 875

Text #4

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2576. Min Act: -0.0000

Data Index: 955228 (C4 (Web Text))

Max Activating Token Index: 312

Text #5

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2555. Min Act: -0.0000

Data Index: 433748 (C4 (Web Text))

Max Activating Token Index: 210

Text #6

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2597. Min Act: -0.0000

Data Index: 637226 (C4 (Web Text))

Max Activating Token Index: 542

Text #7

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2556. Min Act: -0.0000

Data Index: 1303717 (C4 (Web Text))

Max Activating Token Index: 42

Text #8

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2461. Min Act: -0.0000

Data Index: 133347 (C4 (Web Text))

Max Activating Token Index: 607

Text #9

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2438. Min Act: -0.0000

Data Index: 1570690 (Python Code)

Max Activating Token Index: 15

Text #10

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2364. Min Act: -0.0000

Data Index: 355900 (C4 (Web Text))

Max Activating Token Index: 860

Text #11

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2468. Min Act: -0.0000

Data Index: 38698 (C4 (Web Text))

Max Activating Token Index: 542

Text #12

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2576. Min Act: -0.0000

Data Index: 275237 (C4 (Web Text))

Max Activating Token Index: 565

Text #13

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2425. Min Act: -0.0000

Data Index: 1481969 (Python Code)

Max Activating Token Index: 343

Text #14

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2445. Min Act: -0.0000

Data Index: 1079172 (C4 (Web Text))

Max Activating Token Index: 655

Text #15

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2496. Min Act: -0.0000

Data Index: 878450 (C4 (Web Text))

Max Activating Token Index: 921

Text #16

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2337. Min Act: -0.0000

Data Index: 761037 (C4 (Web Text))

Max Activating Token Index: 99

Text #17

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2667. Min Act: -0.0000

Data Index: 729952 (C4 (Web Text))

Max Activating Token Index: 87

Text #18

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2235. Min Act: -0.0000

Data Index: 218062 (C4 (Web Text))

Max Activating Token Index: 129

Text #19

Max Range: 0.3132. Min Range: -0.3132

Max Act: 0.2371. Min Act: -0.0000

Data Index: 90208 (C4 (Web Text))

Max Activating Token Index: 441