Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3745 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.4874. Min Act: -0.0001

Data Index: 676300 (C4 (Web Text))

Max Activating Token Index: 238

Text #1

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.4497. Min Act: -0.0000

Data Index: 406694 (C4 (Web Text))

Max Activating Token Index: 680

Text #2

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.4633. Min Act: -0.0001

Data Index: 371192 (C4 (Web Text))

Max Activating Token Index: 903

Text #3

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3637. Min Act: -0.0001

Data Index: 193005 (C4 (Web Text))

Max Activating Token Index: 361

Text #4

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3574. Min Act: -0.0000

Data Index: 1642607 (Python Code)

Max Activating Token Index: 366

Text #5

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3563. Min Act: -0.0000

Data Index: 342340 (C4 (Web Text))

Max Activating Token Index: 70

Text #6

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3701. Min Act: -0.0000

Data Index: 225709 (C4 (Web Text))

Max Activating Token Index: 210

Text #7

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3401. Min Act: -0.0001

Data Index: 1262603 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3892. Min Act: -0.0000

Data Index: 1121463 (C4 (Web Text))

Max Activating Token Index: 331

Text #9

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3441. Min Act: -0.0000

Data Index: 1149219 (C4 (Web Text))

Max Activating Token Index: 343

Text #10

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3451. Min Act: -0.0000

Data Index: 1345353 (C4 (Web Text))

Max Activating Token Index: 436

Text #11

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3448. Min Act: -0.0001

Data Index: 965947 (C4 (Web Text))

Max Activating Token Index: 195

Text #12

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3343. Min Act: -0.0001

Data Index: 1402514 (Python Code)

Max Activating Token Index: 37

Text #13

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3138. Min Act: -0.0000

Data Index: 371715 (C4 (Web Text))

Max Activating Token Index: 376

Text #14

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3038. Min Act: -0.0000

Data Index: 1320772 (C4 (Web Text))

Max Activating Token Index: 348

Text #15

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3362. Min Act: -0.0000

Data Index: 1321421 (C4 (Web Text))

Max Activating Token Index: 426

Text #16

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3195. Min Act: -0.0000

Data Index: 965702 (C4 (Web Text))

Max Activating Token Index: 1003

Text #17

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3052. Min Act: -0.0000

Data Index: 798145 (C4 (Web Text))

Max Activating Token Index: 270

Text #18

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.3109. Min Act: -0.0001

Data Index: 1067574 (C4 (Web Text))

Max Activating Token Index: 700

Text #19

Max Range: 0.4874. Min Range: -0.4874

Max Act: 0.2867. Min Act: -0.0001

Data Index: 695547 (C4 (Web Text))

Max Activating Token Index: 325