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

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

Neuron 2167 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: 0.5827. Min Range: -0.5827

Max Act: 0.5827. Min Act: -0.0000

Data Index: 225280 (C4 (Web Text))

Max Activating Token Index: 923

Text #1

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4805. Min Act: -0.0000

Data Index: 144804 (C4 (Web Text))

Max Activating Token Index: 31

Text #2

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.5412. Min Act: -0.0000

Data Index: 583581 (C4 (Web Text))

Max Activating Token Index: 15

Text #3

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.5075. Min Act: 0.0000

Data Index: 1274123 (C4 (Web Text))

Max Activating Token Index: 981

Text #4

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.5135. Min Act: -0.0000

Data Index: 203062 (C4 (Web Text))

Max Activating Token Index: 185

Text #5

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4950. Min Act: -0.0000

Data Index: 1185473 (C4 (Web Text))

Max Activating Token Index: 669

Text #6

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.5296. Min Act: -0.0000

Data Index: 55418 (C4 (Web Text))

Max Activating Token Index: 1019

Text #7

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4833. Min Act: -0.0000

Data Index: 1191821 (C4 (Web Text))

Max Activating Token Index: 754

Text #8

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4748. Min Act: -0.0000

Data Index: 1594724 (Python Code)

Max Activating Token Index: 918

Text #9

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4391. Min Act: -0.0000

Data Index: 1114256 (C4 (Web Text))

Max Activating Token Index: 269

Text #10

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4912. Min Act: -0.0000

Data Index: 395596 (C4 (Web Text))

Max Activating Token Index: 277

Text #11

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4817. Min Act: -0.0000

Data Index: 469792 (C4 (Web Text))

Max Activating Token Index: 583

Text #12

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4799. Min Act: -0.0000

Data Index: 1419559 (Python Code)

Max Activating Token Index: 203

Text #13

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4939. Min Act: -0.0000

Data Index: 819555 (C4 (Web Text))

Max Activating Token Index: 728

Text #14

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4525. Min Act: -0.0000

Data Index: 1265677 (C4 (Web Text))

Max Activating Token Index: 466

Text #15

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4557. Min Act: -0.0000

Data Index: 674878 (C4 (Web Text))

Max Activating Token Index: 314

Text #16

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4482. Min Act: -0.0000

Data Index: 584786 (C4 (Web Text))

Max Activating Token Index: 838

Text #17

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4384. Min Act: -0.0000

Data Index: 1396665 (Python Code)

Max Activating Token Index: 237

Text #18

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4594. Min Act: -0.0000

Data Index: 543472 (C4 (Web Text))

Max Activating Token Index: 262

Text #19

Max Range: 0.5827. Min Range: -0.5827

Max Act: 0.4635. Min Act: -0.0000

Data Index: 1572681 (Python Code)

Max Activating Token Index: 796