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

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

Neuron 2460 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0711. Min Range: -7.0711

Max Act: 7.0711. Min Act: -0.0000

Data Index: 709753 (C4 (Web Text))

Max Activating Token Index: 886

Text #1

Max Range: 7.0711. Min Range: -7.0711

Max Act: 6.4581. Min Act: -0.0000

Data Index: 584693 (C4 (Web Text))

Max Activating Token Index: 154

Text #2

Max Range: 7.0711. Min Range: -7.0711

Max Act: 6.2467. Min Act: -0.0000

Data Index: 300879 (C4 (Web Text))

Max Activating Token Index: 39

Text #3

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.9958. Min Act: -0.0000

Data Index: 1231221 (C4 (Web Text))

Max Activating Token Index: 849

Text #4

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.9496. Min Act: -0.0000

Data Index: 879349 (C4 (Web Text))

Max Activating Token Index: 590

Text #5

Max Range: 7.0711. Min Range: -7.0711

Max Act: 6.1347. Min Act: -0.0000

Data Index: 1130966 (C4 (Web Text))

Max Activating Token Index: 945

Text #6

Max Range: 7.0711. Min Range: -7.0711

Max Act: 6.0435. Min Act: -0.0000

Data Index: 1002297 (C4 (Web Text))

Max Activating Token Index: 63

Text #7

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.9985. Min Act: -0.0000

Data Index: 1461884 (Python Code)

Max Activating Token Index: 903

Text #8

Max Range: 7.0711. Min Range: -7.0711

Max Act: 6.1546. Min Act: -0.0000

Data Index: 1436102 (Python Code)

Max Activating Token Index: 971

Text #9

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.5070. Min Act: -0.0000

Data Index: 937490 (C4 (Web Text))

Max Activating Token Index: 777

Text #10

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.8385. Min Act: -0.0000

Data Index: 16500 (C4 (Web Text))

Max Activating Token Index: 983

Text #11

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.8175. Min Act: -0.0000

Data Index: 281369 (C4 (Web Text))

Max Activating Token Index: 174

Text #12

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.7951. Min Act: -0.0000

Data Index: 1444584 (Python Code)

Max Activating Token Index: 795

Text #13

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.7438. Min Act: -0.0000

Data Index: 1342344 (C4 (Web Text))

Max Activating Token Index: 277

Text #14

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.8407. Min Act: -0.0000

Data Index: 297379 (C4 (Web Text))

Max Activating Token Index: 50

Text #15

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.6317. Min Act: -0.0000

Data Index: 3693 (C4 (Web Text))

Max Activating Token Index: 845

Text #16

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.4053. Min Act: -0.0000

Data Index: 1226126 (C4 (Web Text))

Max Activating Token Index: 709

Text #17

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.6264. Min Act: -0.0000

Data Index: 499547 (C4 (Web Text))

Max Activating Token Index: 34

Text #18

Max Range: 7.0711. Min Range: -7.0711

Max Act: 5.3851. Min Act: -0.0000

Data Index: 588764 (C4 (Web Text))

Max Activating Token Index: 55

Text #19

Max Range: 7.0711. Min Range: -7.0711

Max Act: 4.9354. Min Act: -0.0000

Data Index: 1399783 (Python Code)

Max Activating Token Index: 764