Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1475 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4985. Min Act: -0.0001

Data Index: 1465540 (Python Code)

Max Activating Token Index: 200

Text #1

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4999. Min Act: -0.0001

Data Index: 823388 (C4 (Web Text))

Max Activating Token Index: 701

Text #2

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4788. Min Act: -0.0001

Data Index: 67989 (C4 (Web Text))

Max Activating Token Index: 954

Text #3

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4542. Min Act: -0.0001

Data Index: 1062737 (C4 (Web Text))

Max Activating Token Index: 543

Text #4

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4659. Min Act: -0.0002

Data Index: 563462 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4430. Min Act: -0.0002

Data Index: 1255253 (C4 (Web Text))

Max Activating Token Index: 646

Text #6

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4404. Min Act: -0.0002

Data Index: 207143 (C4 (Web Text))

Max Activating Token Index: 733

Text #7

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4443. Min Act: -0.0002

Data Index: 827112 (C4 (Web Text))

Max Activating Token Index: 656

Text #8

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4517. Min Act: -0.0001

Data Index: 1499030 (Python Code)

Max Activating Token Index: 453

Text #9

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4561. Min Act: -0.0002

Data Index: 274967 (C4 (Web Text))

Max Activating Token Index: 818

Text #10

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4507. Min Act: -0.0001

Data Index: 1330074 (C4 (Web Text))

Max Activating Token Index: 98

Text #11

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4434. Min Act: -0.0002

Data Index: 610641 (C4 (Web Text))

Max Activating Token Index: 929

Text #12

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4588. Min Act: -0.0002

Data Index: 2824 (C4 (Web Text))

Max Activating Token Index: 935

Text #13

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4194. Min Act: -0.0001

Data Index: 144041 (C4 (Web Text))

Max Activating Token Index: 339

Text #14

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4472. Min Act: -0.0001

Data Index: 709151 (C4 (Web Text))

Max Activating Token Index: 956

Text #15

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4525. Min Act: -0.0001

Data Index: 775947 (C4 (Web Text))

Max Activating Token Index: 934

Text #16

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4418. Min Act: -0.0001

Data Index: 1461719 (Python Code)

Max Activating Token Index: 871

Text #17

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4300. Min Act: -0.0001

Data Index: 1136145 (C4 (Web Text))

Max Activating Token Index: 372

Text #18

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4415. Min Act: -0.0001

Data Index: 1542537 (Python Code)

Max Activating Token Index: 322

Text #19

Max Range: 0.4999. Min Range: -0.4999

Max Act: 0.4283. Min Act: -0.0001

Data Index: 1241966 (C4 (Web Text))

Max Activating Token Index: 53