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

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

Neuron 4538 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0440. Min Act: -0.0002

Data Index: 799669 (C4 (Web Text))

Max Activating Token Index: 71

Text #1

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0430. Min Act: -0.0002

Data Index: 578124 (C4 (Web Text))

Max Activating Token Index: 543

Text #2

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0418. Min Act: -0.0002

Data Index: 238325 (C4 (Web Text))

Max Activating Token Index: 383

Text #3

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0400. Min Act: -0.0002

Data Index: 1341317 (C4 (Web Text))

Max Activating Token Index: 137

Text #4

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0400. Min Act: -0.0002

Data Index: 753009 (C4 (Web Text))

Max Activating Token Index: 267

Text #5

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0386. Min Act: -0.0002

Data Index: 1274213 (C4 (Web Text))

Max Activating Token Index: 902

Text #6

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0388. Min Act: -0.0002

Data Index: 1184775 (C4 (Web Text))

Max Activating Token Index: 333

Text #7

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0377. Min Act: -0.0002

Data Index: 790065 (C4 (Web Text))

Max Activating Token Index: 331

Text #8

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0335. Min Act: -0.0002

Data Index: 1404843 (Python Code)

Max Activating Token Index: 67

Text #9

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0358. Min Act: -0.0002

Data Index: 1320942 (C4 (Web Text))

Max Activating Token Index: 913

Text #10

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0330. Min Act: -0.0002

Data Index: 760714 (C4 (Web Text))

Max Activating Token Index: 488

Text #11

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0311. Min Act: -0.0002

Data Index: 1417439 (Python Code)

Max Activating Token Index: 792

Text #12

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0350. Min Act: -0.0002

Data Index: 249452 (C4 (Web Text))

Max Activating Token Index: 658

Text #13

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0340. Min Act: -0.0002

Data Index: 995440 (C4 (Web Text))

Max Activating Token Index: 387

Text #14

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0311. Min Act: -0.0002

Data Index: 501475 (C4 (Web Text))

Max Activating Token Index: 274

Text #15

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0293. Min Act: -0.0002

Data Index: 480811 (C4 (Web Text))

Max Activating Token Index: 595

Text #16

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0324. Min Act: -0.0002

Data Index: 1275756 (C4 (Web Text))

Max Activating Token Index: 248

Text #17

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0332. Min Act: -0.0002

Data Index: 1480203 (Python Code)

Max Activating Token Index: 57

Text #18

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0322. Min Act: -0.0002

Data Index: 520490 (C4 (Web Text))

Max Activating Token Index: 190

Text #19

Max Range: 0.0440. Min Range: -0.0440

Max Act: 0.0292. Min Act: -0.0002

Data Index: 661851 (C4 (Web Text))

Max Activating Token Index: 654