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

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

Neuron 500 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.5915. Min Act: -0.0000

Data Index: 1558386 (Python Code)

Max Activating Token Index: 568

Text #1

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.7553. Min Act: -0.0000

Data Index: 1576450 (Python Code)

Max Activating Token Index: 765

Text #2

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.2788. Min Act: -0.0000

Data Index: 1408697 (Python Code)

Max Activating Token Index: 633

Text #3

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.7028. Min Act: -0.0000

Data Index: 1554770 (Python Code)

Max Activating Token Index: 590

Text #4

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.6023. Min Act: -0.0000

Data Index: 165561 (C4 (Web Text))

Max Activating Token Index: 753

Text #5

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.4098. Min Act: -0.0000

Data Index: 1643456 (Python Code)

Max Activating Token Index: 734

Text #6

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.4084. Min Act: -0.0000

Data Index: 1470380 (Python Code)

Max Activating Token Index: 829

Text #7

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.3825. Min Act: -0.0000

Data Index: 313929 (C4 (Web Text))

Max Activating Token Index: 814

Text #8

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.3649. Min Act: -0.0000

Data Index: 1542485 (Python Code)

Max Activating Token Index: 152

Text #9

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.5288. Min Act: -0.0000

Data Index: 234129 (C4 (Web Text))

Max Activating Token Index: 879

Text #10

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.3353. Min Act: -0.0000

Data Index: 691908 (C4 (Web Text))

Max Activating Token Index: 521

Text #11

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.1972. Min Act: -0.0000

Data Index: 1550248 (Python Code)

Max Activating Token Index: 191

Text #12

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.3291. Min Act: -0.0000

Data Index: 455972 (C4 (Web Text))

Max Activating Token Index: 166

Text #13

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.3202. Min Act: -0.0000

Data Index: 1657091 (Python Code)

Max Activating Token Index: 630

Text #14

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.1877. Min Act: -0.0000

Data Index: 1650203 (Python Code)

Max Activating Token Index: 473

Text #15

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.4596. Min Act: -0.0001

Data Index: 1056609 (C4 (Web Text))

Max Activating Token Index: 108

Text #16

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.1412. Min Act: -0.0000

Data Index: 1593751 (Python Code)

Max Activating Token Index: 19

Text #17

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.4329. Min Act: -0.0000

Data Index: 1370278 (Python Code)

Max Activating Token Index: 244

Text #18

Max Range: 3.7553. Min Range: -3.7553

Max Act: 3.2896. Min Act: -0.0000

Data Index: 1451091 (Python Code)

Max Activating Token Index: 666

Text #19

Max Range: 3.7553. Min Range: -3.7553

Max Act: 2.9413. Min Act: -0.0000

Data Index: 104160 (C4 (Web Text))

Max Activating Token Index: 560