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

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

Neuron 3597 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.4432. Min Act: -0.0001

Data Index: 1475440 (Python Code)

Max Activating Token Index: 436

Text #1

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.4208. Min Act: -0.0001

Data Index: 235938 (C4 (Web Text))

Max Activating Token Index: 293

Text #2

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3186. Min Act: -0.0001

Data Index: 1197439 (C4 (Web Text))

Max Activating Token Index: 275

Text #3

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3631. Min Act: -0.0001

Data Index: 421580 (C4 (Web Text))

Max Activating Token Index: 778

Text #4

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3290. Min Act: -0.0001

Data Index: 637337 (C4 (Web Text))

Max Activating Token Index: 981

Text #5

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3544. Min Act: -0.0001

Data Index: 907498 (C4 (Web Text))

Max Activating Token Index: 556

Text #6

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.4143. Min Act: -0.0001

Data Index: 1045973 (C4 (Web Text))

Max Activating Token Index: 146

Text #7

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3542. Min Act: -0.0001

Data Index: 121368 (C4 (Web Text))

Max Activating Token Index: 475

Text #8

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3136. Min Act: -0.0001

Data Index: 497691 (C4 (Web Text))

Max Activating Token Index: 771

Text #9

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3510. Min Act: -0.0001

Data Index: 1310761 (C4 (Web Text))

Max Activating Token Index: 412

Text #10

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3108. Min Act: -0.0001

Data Index: 622439 (C4 (Web Text))

Max Activating Token Index: 727

Text #11

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3407. Min Act: -0.0001

Data Index: 833680 (C4 (Web Text))

Max Activating Token Index: 595

Text #12

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3255. Min Act: -0.0001

Data Index: 1111772 (C4 (Web Text))

Max Activating Token Index: 738

Text #13

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3233. Min Act: -0.0001

Data Index: 1115473 (C4 (Web Text))

Max Activating Token Index: 25

Text #14

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3237. Min Act: -0.0001

Data Index: 172573 (C4 (Web Text))

Max Activating Token Index: 120

Text #15

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3497. Min Act: -0.0001

Data Index: 288777 (C4 (Web Text))

Max Activating Token Index: 710

Text #16

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3183. Min Act: -0.0001

Data Index: 868939 (C4 (Web Text))

Max Activating Token Index: 658

Text #17

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3579. Min Act: -0.0001

Data Index: 379887 (C4 (Web Text))

Max Activating Token Index: 784

Text #18

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3624. Min Act: -0.0001

Data Index: 1269301 (C4 (Web Text))

Max Activating Token Index: 835

Text #19

Max Range: 0.4432. Min Range: -0.4432

Max Act: 0.3270. Min Act: -0.0001

Data Index: 1058019 (C4 (Web Text))

Max Activating Token Index: 881