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

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

Neuron 4910 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: 1.8506. Min Range: -1.8506

Max Act: 1.8268. Min Act: -0.0000

Data Index: 538821 (C4 (Web Text))

Max Activating Token Index: 744

Text #1

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.8506. Min Act: -0.0000

Data Index: 1163929 (C4 (Web Text))

Max Activating Token Index: 341

Text #2

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.7297. Min Act: -0.0000

Data Index: 1373388 (Python Code)

Max Activating Token Index: 989

Text #3

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.5529. Min Act: -0.0000

Data Index: 988917 (C4 (Web Text))

Max Activating Token Index: 70

Text #4

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.5199. Min Act: -0.0000

Data Index: 523792 (C4 (Web Text))

Max Activating Token Index: 351

Text #5

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.5172. Min Act: -0.0000

Data Index: 1307315 (C4 (Web Text))

Max Activating Token Index: 719

Text #6

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.4758. Min Act: -0.0000

Data Index: 155736 (C4 (Web Text))

Max Activating Token Index: 32

Text #7

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.4599. Min Act: -0.0000

Data Index: 396709 (C4 (Web Text))

Max Activating Token Index: 541

Text #8

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.4653. Min Act: -0.0000

Data Index: 1530522 (Python Code)

Max Activating Token Index: 1003

Text #9

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.4254. Min Act: -0.0000

Data Index: 1040290 (C4 (Web Text))

Max Activating Token Index: 903

Text #10

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.5174. Min Act: -0.0000

Data Index: 708154 (C4 (Web Text))

Max Activating Token Index: 100

Text #11

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.5632. Min Act: -0.0000

Data Index: 1432480 (Python Code)

Max Activating Token Index: 290

Text #12

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.3567. Min Act: -0.0000

Data Index: 981398 (C4 (Web Text))

Max Activating Token Index: 202

Text #13

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.2959. Min Act: -0.0000

Data Index: 1415254 (Python Code)

Max Activating Token Index: 234

Text #14

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.2973. Min Act: -0.0000

Data Index: 1420025 (Python Code)

Max Activating Token Index: 117

Text #15

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.4171. Min Act: -0.0000

Data Index: 474735 (C4 (Web Text))

Max Activating Token Index: 221

Text #16

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.2937. Min Act: -0.0000

Data Index: 551486 (C4 (Web Text))

Max Activating Token Index: 864

Text #17

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.3613. Min Act: -0.0000

Data Index: 649273 (C4 (Web Text))

Max Activating Token Index: 540

Text #18

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.3508. Min Act: -0.0000

Data Index: 1127047 (C4 (Web Text))

Max Activating Token Index: 545

Text #19

Max Range: 1.8506. Min Range: -1.8506

Max Act: 1.2359. Min Act: -0.0000

Data Index: 467133 (C4 (Web Text))

Max Activating Token Index: 108