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

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

Neuron 2339 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: 0.3897. Min Range: -0.3897

Max Act: 0.3820. Min Act: -0.0000

Data Index: 1356928 (C4 (Web Text))

Max Activating Token Index: 434

Text #1

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3897. Min Act: -0.0000

Data Index: 1404353 (Python Code)

Max Activating Token Index: 72

Text #2

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3158. Min Act: -0.0000

Data Index: 344290 (C4 (Web Text))

Max Activating Token Index: 581

Text #3

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3311. Min Act: -0.0000

Data Index: 255155 (C4 (Web Text))

Max Activating Token Index: 627

Text #4

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3633. Min Act: -0.0000

Data Index: 936928 (C4 (Web Text))

Max Activating Token Index: 347

Text #5

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3326. Min Act: -0.0000

Data Index: 172063 (C4 (Web Text))

Max Activating Token Index: 860

Text #6

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2924. Min Act: -0.0000

Data Index: 846614 (C4 (Web Text))

Max Activating Token Index: 608

Text #7

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3329. Min Act: -0.0000

Data Index: 404181 (C4 (Web Text))

Max Activating Token Index: 43

Text #8

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3276. Min Act: -0.0000

Data Index: 503025 (C4 (Web Text))

Max Activating Token Index: 280

Text #9

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3296. Min Act: -0.0000

Data Index: 57288 (C4 (Web Text))

Max Activating Token Index: 525

Text #10

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3227. Min Act: -0.0000

Data Index: 1362939 (Python Code)

Max Activating Token Index: 181

Text #11

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2982. Min Act: -0.0000

Data Index: 684848 (C4 (Web Text))

Max Activating Token Index: 578

Text #12

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.3022. Min Act: -0.0000

Data Index: 1565135 (Python Code)

Max Activating Token Index: 963

Text #13

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2924. Min Act: -0.0000

Data Index: 1573717 (Python Code)

Max Activating Token Index: 967

Text #14

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2917. Min Act: -0.0000

Data Index: 703947 (C4 (Web Text))

Max Activating Token Index: 328

Text #15

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2962. Min Act: -0.0000

Data Index: 1412748 (Python Code)

Max Activating Token Index: 335

Text #16

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2875. Min Act: -0.0000

Data Index: 570830 (C4 (Web Text))

Max Activating Token Index: 268

Text #17

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2758. Min Act: -0.0000

Data Index: 697288 (C4 (Web Text))

Max Activating Token Index: 570

Text #18

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2723. Min Act: -0.0000

Data Index: 1136225 (C4 (Web Text))

Max Activating Token Index: 676

Text #19

Max Range: 0.3897. Min Range: -0.3897

Max Act: 0.2885. Min Act: -0.0000

Data Index: 646376 (C4 (Web Text))

Max Activating Token Index: 473