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

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

Neuron 3581 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.4309. Min Act: -0.0000

Data Index: 1399103 (Python Code)

Max Activating Token Index: 606

Text #1

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.4575. Min Act: -0.0000

Data Index: 1279079 (C4 (Web Text))

Max Activating Token Index: 318

Text #2

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.4171. Min Act: -0.0000

Data Index: 1095373 (C4 (Web Text))

Max Activating Token Index: 892

Text #3

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3795. Min Act: -0.0000

Data Index: 1501751 (Python Code)

Max Activating Token Index: 665

Text #4

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3901. Min Act: -0.0000

Data Index: 1435522 (Python Code)

Max Activating Token Index: 492

Text #5

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3810. Min Act: -0.0000

Data Index: 762543 (C4 (Web Text))

Max Activating Token Index: 36

Text #6

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3710. Min Act: -0.0000

Data Index: 479450 (C4 (Web Text))

Max Activating Token Index: 420

Text #7

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3687. Min Act: -0.0000

Data Index: 1397721 (Python Code)

Max Activating Token Index: 431

Text #8

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3592. Min Act: 0.0000

Data Index: 1497237 (Python Code)

Max Activating Token Index: 235

Text #9

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3616. Min Act: -0.0000

Data Index: 1399741 (Python Code)

Max Activating Token Index: 522

Text #10

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3615. Min Act: -0.0000

Data Index: 1476947 (Python Code)

Max Activating Token Index: 661

Text #11

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3565. Min Act: -0.0000

Data Index: 1109360 (C4 (Web Text))

Max Activating Token Index: 621

Text #12

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3391. Min Act: -0.0000

Data Index: 1462616 (Python Code)

Max Activating Token Index: 642

Text #13

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3277. Min Act: 0.0000

Data Index: 1500052 (Python Code)

Max Activating Token Index: 480

Text #14

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3378. Min Act: -0.0000

Data Index: 181141 (C4 (Web Text))

Max Activating Token Index: 480

Text #15

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3248. Min Act: -0.0000

Data Index: 298405 (C4 (Web Text))

Max Activating Token Index: 913

Text #16

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3350. Min Act: -0.0000

Data Index: 1533296 (Python Code)

Max Activating Token Index: 766

Text #17

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3428. Min Act: -0.0000

Data Index: 1455474 (Python Code)

Max Activating Token Index: 1020

Text #18

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3322. Min Act: -0.0000

Data Index: 868924 (C4 (Web Text))

Max Activating Token Index: 405

Text #19

Max Range: 0.4575. Min Range: -0.4575

Max Act: 0.3231. Min Act: -0.0000

Data Index: 1167875 (C4 (Web Text))

Max Activating Token Index: 264