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

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

Neuron 4080 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: 2.9475. Min Range: -2.9475

Max Act: 2.9475. Min Act: -0.0000

Data Index: 765775 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.6842. Min Act: -0.0000

Data Index: 147151 (C4 (Web Text))

Max Activating Token Index: 700

Text #2

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.4340. Min Act: -0.0000

Data Index: 656829 (C4 (Web Text))

Max Activating Token Index: 246

Text #3

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.3235. Min Act: -0.0000

Data Index: 1402940 (Python Code)

Max Activating Token Index: 937

Text #4

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.3019. Min Act: -0.0000

Data Index: 1512058 (Python Code)

Max Activating Token Index: 16

Text #5

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.1860. Min Act: -0.0000

Data Index: 662394 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.1083. Min Act: -0.0000

Data Index: 224464 (C4 (Web Text))

Max Activating Token Index: 575

Text #7

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.1116. Min Act: -0.0000

Data Index: 1130219 (C4 (Web Text))

Max Activating Token Index: 418

Text #8

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.1010. Min Act: -0.0000

Data Index: 810739 (C4 (Web Text))

Max Activating Token Index: 92

Text #9

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.0919. Min Act: -0.0000

Data Index: 1454145 (Python Code)

Max Activating Token Index: 242

Text #10

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.0540. Min Act: -0.0000

Data Index: 369631 (C4 (Web Text))

Max Activating Token Index: 490

Text #11

Max Range: 2.9475. Min Range: -2.9475

Max Act: 2.0660. Min Act: -0.0000

Data Index: 1255622 (C4 (Web Text))

Max Activating Token Index: 243

Text #12

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8977. Min Act: -0.0000

Data Index: 1027987 (C4 (Web Text))

Max Activating Token Index: 133

Text #13

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.9851. Min Act: -0.0000

Data Index: 922268 (C4 (Web Text))

Max Activating Token Index: 546

Text #14

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8760. Min Act: -0.0000

Data Index: 315110 (C4 (Web Text))

Max Activating Token Index: 215

Text #15

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8574. Min Act: -0.0000

Data Index: 1244062 (C4 (Web Text))

Max Activating Token Index: 1006

Text #16

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8598. Min Act: -0.0000

Data Index: 959884 (C4 (Web Text))

Max Activating Token Index: 826

Text #17

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8442. Min Act: -0.0000

Data Index: 153567 (C4 (Web Text))

Max Activating Token Index: 983

Text #18

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.8504. Min Act: -0.0000

Data Index: 1348177 (C4 (Web Text))

Max Activating Token Index: 814

Text #19

Max Range: 2.9475. Min Range: -2.9475

Max Act: 1.9130. Min Act: -0.0000

Data Index: 868415 (C4 (Web Text))

Max Activating Token Index: 355