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

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

Neuron 256 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: 0.0724. Min Range: -0.0724

Max Act: 0.0724. Min Act: -0.0000

Data Index: 1006652 (C4 (Web Text))

Max Activating Token Index: 525

Text #1

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0697. Min Act: -0.0001

Data Index: 106470 (C4 (Web Text))

Max Activating Token Index: 436

Text #2

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0657. Min Act: -0.0000

Data Index: 1275114 (C4 (Web Text))

Max Activating Token Index: 900

Text #3

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0625. Min Act: -0.0001

Data Index: 927316 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0711. Min Act: -0.0000

Data Index: 218286 (C4 (Web Text))

Max Activating Token Index: 965

Text #5

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0593. Min Act: -0.0000

Data Index: 518512 (C4 (Web Text))

Max Activating Token Index: 316

Text #6

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0628. Min Act: -0.0000

Data Index: 327667 (C4 (Web Text))

Max Activating Token Index: 282

Text #7

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0586. Min Act: -0.0000

Data Index: 951689 (C4 (Web Text))

Max Activating Token Index: 643

Text #8

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0613. Min Act: -0.0000

Data Index: 765044 (C4 (Web Text))

Max Activating Token Index: 708

Text #9

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0572. Min Act: -0.0000

Data Index: 997964 (C4 (Web Text))

Max Activating Token Index: 81

Text #10

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0540. Min Act: -0.0000

Data Index: 480079 (C4 (Web Text))

Max Activating Token Index: 210

Text #11

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0607. Min Act: -0.0000

Data Index: 163725 (C4 (Web Text))

Max Activating Token Index: 919

Text #12

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0547. Min Act: -0.0000

Data Index: 1483293 (Python Code)

Max Activating Token Index: 183

Text #13

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0552. Min Act: -0.0000

Data Index: 109097 (C4 (Web Text))

Max Activating Token Index: 1000

Text #14

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0579. Min Act: -0.0000

Data Index: 145735 (C4 (Web Text))

Max Activating Token Index: 774

Text #15

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0548. Min Act: -0.0000

Data Index: 451319 (C4 (Web Text))

Max Activating Token Index: 792

Text #16

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0555. Min Act: -0.0000

Data Index: 1084045 (C4 (Web Text))

Max Activating Token Index: 859

Text #17

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0599. Min Act: -0.0000

Data Index: 318249 (C4 (Web Text))

Max Activating Token Index: 587

Text #18

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0557. Min Act: -0.0001

Data Index: 1118279 (C4 (Web Text))

Max Activating Token Index: 638

Text #19

Max Range: 0.0724. Min Range: -0.0724

Max Act: 0.0569. Min Act: -0.0000

Data Index: 218182 (C4 (Web Text))

Max Activating Token Index: 913