Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1944 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.2605. Min Act: -0.0000

Data Index: 195761 (C4 (Web Text))

Max Activating Token Index: 601

Text #1

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.2063. Min Act: -0.0000

Data Index: 264594 (C4 (Web Text))

Max Activating Token Index: 353

Text #2

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.2071. Min Act: -0.0000

Data Index: 942368 (C4 (Web Text))

Max Activating Token Index: 827

Text #3

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1900. Min Act: -0.0000

Data Index: 354627 (C4 (Web Text))

Max Activating Token Index: 115

Text #4

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1810. Min Act: -0.0000

Data Index: 538221 (C4 (Web Text))

Max Activating Token Index: 336

Text #5

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1733. Min Act: -0.0000

Data Index: 514675 (C4 (Web Text))

Max Activating Token Index: 412

Text #6

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1743. Min Act: -0.0000

Data Index: 486250 (C4 (Web Text))

Max Activating Token Index: 251

Text #7

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1597. Min Act: -0.0000

Data Index: 1390326 (Python Code)

Max Activating Token Index: 323

Text #8

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1730. Min Act: -0.0000

Data Index: 508676 (C4 (Web Text))

Max Activating Token Index: 132

Text #9

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1732. Min Act: -0.0000

Data Index: 592223 (C4 (Web Text))

Max Activating Token Index: 353

Text #10

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1626. Min Act: -0.0000

Data Index: 486382 (C4 (Web Text))

Max Activating Token Index: 3

Text #11

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1521. Min Act: -0.0000

Data Index: 1362753 (Python Code)

Max Activating Token Index: 788

Text #12

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1644. Min Act: -0.0000

Data Index: 852282 (C4 (Web Text))

Max Activating Token Index: 718

Text #13

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1583. Min Act: -0.0000

Data Index: 541860 (C4 (Web Text))

Max Activating Token Index: 238

Text #14

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1542. Min Act: -0.0000

Data Index: 357303 (C4 (Web Text))

Max Activating Token Index: 963

Text #15

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1560. Min Act: -0.0000

Data Index: 267453 (C4 (Web Text))

Max Activating Token Index: 798

Text #16

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1461. Min Act: -0.0000

Data Index: 536447 (C4 (Web Text))

Max Activating Token Index: 949

Text #17

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1668. Min Act: -0.0000

Data Index: 1177867 (C4 (Web Text))

Max Activating Token Index: 336

Text #18

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1562. Min Act: -0.0000

Data Index: 55967 (C4 (Web Text))

Max Activating Token Index: 201

Text #19

Max Range: 0.2605. Min Range: -0.2605

Max Act: 0.1524. Min Act: -0.0000

Data Index: 1412760 (Python Code)

Max Activating Token Index: 844