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

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

Neuron 3870 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0029. Min Act: -0.0000

Data Index: 1449158 (Python Code)

Max Activating Token Index: 235

Text #1

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0028. Min Act: -0.0000

Data Index: 191195 (C4 (Web Text))

Max Activating Token Index: 258

Text #2

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0026. Min Act: -0.0000

Data Index: 1550132 (Python Code)

Max Activating Token Index: 298

Text #3

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0021. Min Act: -0.0000

Data Index: 315481 (C4 (Web Text))

Max Activating Token Index: 867

Text #4

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0031. Min Act: -0.0000

Data Index: 1620477 (Python Code)

Max Activating Token Index: 249

Text #5

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0025. Min Act: -0.0000

Data Index: 938370 (C4 (Web Text))

Max Activating Token Index: 235

Text #6

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0023. Min Act: -0.0000

Data Index: 356025 (C4 (Web Text))

Max Activating Token Index: 297

Text #7

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0020. Min Act: -0.0000

Data Index: 176641 (C4 (Web Text))

Max Activating Token Index: 170

Text #8

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0019. Min Act: -0.0000

Data Index: 1496438 (Python Code)

Max Activating Token Index: 519

Text #9

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0019. Min Act: -0.0001

Data Index: 1460402 (Python Code)

Max Activating Token Index: 597

Text #10

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0020. Min Act: -0.0000

Data Index: 652947 (C4 (Web Text))

Max Activating Token Index: 801

Text #11

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0019. Min Act: -0.0000

Data Index: 347448 (C4 (Web Text))

Max Activating Token Index: 363

Text #12

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0017. Min Act: -0.0000

Data Index: 176365 (C4 (Web Text))

Max Activating Token Index: 136

Text #13

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0020. Min Act: -0.0000

Data Index: 1541646 (Python Code)

Max Activating Token Index: 550

Text #14

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0017. Min Act: -0.0000

Data Index: 1379202 (Python Code)

Max Activating Token Index: 509

Text #15

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0019. Min Act: -0.0000

Data Index: 875841 (C4 (Web Text))

Max Activating Token Index: 579

Text #16

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0016. Min Act: -0.0000

Data Index: 799955 (C4 (Web Text))

Max Activating Token Index: 340

Text #17

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0016. Min Act: -0.0000

Data Index: 1461144 (Python Code)

Max Activating Token Index: 616

Text #18

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0015. Min Act: -0.0000

Data Index: 1123736 (C4 (Web Text))

Max Activating Token Index: 67

Text #19

Max Range: 0.0031. Min Range: -0.0031

Max Act: 0.0014. Min Act: -0.0000

Data Index: 1026775 (C4 (Web Text))

Max Activating Token Index: 230