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

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

Neuron 3273 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: 0.5797. Min Range: -0.5797

Max Act: 0.5797. Min Act: 0.0000

Data Index: 1367307 (Python Code)

Max Activating Token Index: 815

Text #1

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5531. Min Act: -0.0000

Data Index: 97884 (C4 (Web Text))

Max Activating Token Index: 916

Text #2

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5356. Min Act: -0.0000

Data Index: 1456610 (Python Code)

Max Activating Token Index: 379

Text #3

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5275. Min Act: 0.0000

Data Index: 1501585 (Python Code)

Max Activating Token Index: 442

Text #4

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5552. Min Act: -0.0000

Data Index: 1368724 (Python Code)

Max Activating Token Index: 276

Text #5

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5302. Min Act: -0.0000

Data Index: 1176328 (C4 (Web Text))

Max Activating Token Index: 162

Text #6

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5288. Min Act: 0.0000

Data Index: 1501528 (Python Code)

Max Activating Token Index: 744

Text #7

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4902. Min Act: 0.0000

Data Index: 1459486 (Python Code)

Max Activating Token Index: 663

Text #8

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.5027. Min Act: -0.0000

Data Index: 1468704 (Python Code)

Max Activating Token Index: 576

Text #9

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4949. Min Act: -0.0000

Data Index: 357947 (C4 (Web Text))

Max Activating Token Index: 107

Text #10

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4595. Min Act: -0.0000

Data Index: 569552 (C4 (Web Text))

Max Activating Token Index: 741

Text #11

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4773. Min Act: -0.0000

Data Index: 1523494 (Python Code)

Max Activating Token Index: 48

Text #12

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4448. Min Act: 0.0000

Data Index: 1404924 (Python Code)

Max Activating Token Index: 628

Text #13

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4780. Min Act: -0.0000

Data Index: 1587870 (Python Code)

Max Activating Token Index: 960

Text #14

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4462. Min Act: -0.0000

Data Index: 433253 (C4 (Web Text))

Max Activating Token Index: 238

Text #15

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4709. Min Act: -0.0000

Data Index: 397250 (C4 (Web Text))

Max Activating Token Index: 493

Text #16

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4540. Min Act: -0.0000

Data Index: 1643936 (Python Code)

Max Activating Token Index: 528

Text #17

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4377. Min Act: -0.0000

Data Index: 1361015 (Python Code)

Max Activating Token Index: 742

Text #18

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4522. Min Act: 0.0000

Data Index: 1410132 (Python Code)

Max Activating Token Index: 836

Text #19

Max Range: 0.5797. Min Range: -0.5797

Max Act: 0.4449. Min Act: 0.0002

Data Index: 231964 (C4 (Web Text))

Max Activating Token Index: 102