Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 43 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.8741. Min Act: -0.0000

Data Index: 23005 (C4 (Web Text))

Max Activating Token Index: 679

Text #1

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.7967. Min Act: -0.0000

Data Index: 1046540 (C4 (Web Text))

Max Activating Token Index: 992

Text #2

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6777. Min Act: -0.0000

Data Index: 1600544 (Python Code)

Max Activating Token Index: 7

Text #3

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6251. Min Act: -0.0000

Data Index: 1465094 (Python Code)

Max Activating Token Index: 327

Text #4

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6050. Min Act: -0.0000

Data Index: 528161 (C4 (Web Text))

Max Activating Token Index: 458

Text #5

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6253. Min Act: -0.0000

Data Index: 1155997 (C4 (Web Text))

Max Activating Token Index: 642

Text #6

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6076. Min Act: -0.0000

Data Index: 367117 (C4 (Web Text))

Max Activating Token Index: 213

Text #7

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5581. Min Act: -0.0000

Data Index: 792186 (C4 (Web Text))

Max Activating Token Index: 269

Text #8

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.6119. Min Act: -0.0000

Data Index: 1205680 (C4 (Web Text))

Max Activating Token Index: 532

Text #9

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5786. Min Act: -0.0000

Data Index: 395720 (C4 (Web Text))

Max Activating Token Index: 18

Text #10

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5849. Min Act: -0.0000

Data Index: 1399822 (Python Code)

Max Activating Token Index: 669

Text #11

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5675. Min Act: -0.0000

Data Index: 1550794 (Python Code)

Max Activating Token Index: 30

Text #12

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5665. Min Act: -0.0000

Data Index: 1292916 (C4 (Web Text))

Max Activating Token Index: 419

Text #13

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5665. Min Act: -0.0000

Data Index: 1408830 (Python Code)

Max Activating Token Index: 1015

Text #14

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5402. Min Act: -0.0000

Data Index: 1107659 (C4 (Web Text))

Max Activating Token Index: 848

Text #15

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5748. Min Act: -0.0000

Data Index: 1599287 (Python Code)

Max Activating Token Index: 250

Text #16

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5714. Min Act: -0.0000

Data Index: 1412583 (Python Code)

Max Activating Token Index: 6

Text #17

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5347. Min Act: -0.0000

Data Index: 51640 (C4 (Web Text))

Max Activating Token Index: 538

Text #18

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5593. Min Act: -0.0000

Data Index: 1314020 (C4 (Web Text))

Max Activating Token Index: 631

Text #19

Max Range: 0.8741. Min Range: -0.8741

Max Act: 0.5107. Min Act: -0.0000

Data Index: 215152 (C4 (Web Text))

Max Activating Token Index: 395