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

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

Neuron 553 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: 3.5018. Min Range: -3.5018

Max Act: 3.5018. Min Act: -0.0000

Data Index: 4116 (C4 (Web Text))

Max Activating Token Index: 820

Text #1

Max Range: 3.5018. Min Range: -3.5018

Max Act: 3.0583. Min Act: -0.0000

Data Index: 743920 (C4 (Web Text))

Max Activating Token Index: 912

Text #2

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3993. Min Act: -0.0000

Data Index: 1381536 (Python Code)

Max Activating Token Index: 700

Text #3

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.5041. Min Act: -0.0000

Data Index: 525089 (C4 (Web Text))

Max Activating Token Index: 933

Text #4

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3493. Min Act: -0.0000

Data Index: 1501030 (Python Code)

Max Activating Token Index: 858

Text #5

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.4110. Min Act: -0.0000

Data Index: 1550551 (Python Code)

Max Activating Token Index: 798

Text #6

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3999. Min Act: -0.0000

Data Index: 898444 (C4 (Web Text))

Max Activating Token Index: 717

Text #7

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3853. Min Act: -0.0000

Data Index: 1428972 (Python Code)

Max Activating Token Index: 677

Text #8

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3569. Min Act: -0.0000

Data Index: 36668 (C4 (Web Text))

Max Activating Token Index: 872

Text #9

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.2327. Min Act: -0.0000

Data Index: 205563 (C4 (Web Text))

Max Activating Token Index: 567

Text #10

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3427. Min Act: -0.0000

Data Index: 1291299 (C4 (Web Text))

Max Activating Token Index: 509

Text #11

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.2175. Min Act: -0.0000

Data Index: 416561 (C4 (Web Text))

Max Activating Token Index: 678

Text #12

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3333. Min Act: -0.0000

Data Index: 805091 (C4 (Web Text))

Max Activating Token Index: 855

Text #13

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3303. Min Act: -0.0000

Data Index: 454988 (C4 (Web Text))

Max Activating Token Index: 448

Text #14

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.1882. Min Act: -0.0000

Data Index: 1606713 (Python Code)

Max Activating Token Index: 24

Text #15

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.1919. Min Act: -0.0000

Data Index: 120823 (C4 (Web Text))

Max Activating Token Index: 155

Text #16

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.1878. Min Act: -0.0000

Data Index: 1028182 (C4 (Web Text))

Max Activating Token Index: 355

Text #17

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.1712. Min Act: -0.0000

Data Index: 1610435 (Python Code)

Max Activating Token Index: 602

Text #18

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.2723. Min Act: -0.0000

Data Index: 1206864 (C4 (Web Text))

Max Activating Token Index: 473

Text #19

Max Range: 3.5018. Min Range: -3.5018

Max Act: 2.3574. Min Act: -0.0000

Data Index: 191779 (C4 (Web Text))

Max Activating Token Index: 423