Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1730 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2646. Min Act: -0.0001

Data Index: 1354944 (C4 (Web Text))

Max Activating Token Index: 293

Text #1

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2650. Min Act: -0.0001

Data Index: 665151 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2343. Min Act: -0.0001

Data Index: 642479 (C4 (Web Text))

Max Activating Token Index: 429

Text #3

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2261. Min Act: -0.0001

Data Index: 920132 (C4 (Web Text))

Max Activating Token Index: 48

Text #4

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2233. Min Act: -0.0001

Data Index: 1067186 (C4 (Web Text))

Max Activating Token Index: 551

Text #5

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2358. Min Act: -0.0001

Data Index: 665928 (C4 (Web Text))

Max Activating Token Index: 300

Text #6

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2206. Min Act: -0.0001

Data Index: 578696 (C4 (Web Text))

Max Activating Token Index: 691

Text #7

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1796. Min Act: -0.0001

Data Index: 879171 (C4 (Web Text))

Max Activating Token Index: 869

Text #8

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1900. Min Act: -0.0001

Data Index: 1512752 (Python Code)

Max Activating Token Index: 634

Text #9

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2077. Min Act: -0.0001

Data Index: 959373 (C4 (Web Text))

Max Activating Token Index: 196

Text #10

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.2044. Min Act: -0.0001

Data Index: 918306 (C4 (Web Text))

Max Activating Token Index: 349

Text #11

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1971. Min Act: -0.0001

Data Index: 969713 (C4 (Web Text))

Max Activating Token Index: 669

Text #12

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1956. Min Act: -0.0001

Data Index: 455399 (C4 (Web Text))

Max Activating Token Index: 861

Text #13

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1849. Min Act: -0.0001

Data Index: 383337 (C4 (Web Text))

Max Activating Token Index: 777

Text #14

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1923. Min Act: -0.0001

Data Index: 475472 (C4 (Web Text))

Max Activating Token Index: 662

Text #15

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1987. Min Act: -0.0001

Data Index: 464795 (C4 (Web Text))

Max Activating Token Index: 46

Text #16

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1918. Min Act: -0.0001

Data Index: 1422430 (Python Code)

Max Activating Token Index: 600

Text #17

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1758. Min Act: -0.0001

Data Index: 1211484 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1867. Min Act: -0.0001

Data Index: 490326 (C4 (Web Text))

Max Activating Token Index: 1000

Text #19

Max Range: 0.2650. Min Range: -0.2650

Max Act: 0.1864. Min Act: -0.0001

Data Index: 526181 (C4 (Web Text))

Max Activating Token Index: 79