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

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

Neuron 5003 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5854. Min Range: -2.5854

Max Act: 2.5341. Min Act: -0.0001

Data Index: 1035122 (C4 (Web Text))

Max Activating Token Index: 859

Text #1

Max Range: 2.5854. Min Range: -2.5854

Max Act: 2.5854. Min Act: -0.0000

Data Index: 474865 (C4 (Web Text))

Max Activating Token Index: 206

Text #2

Max Range: 2.5854. Min Range: -2.5854

Max Act: 2.3305. Min Act: -0.0000

Data Index: 583558 (C4 (Web Text))

Max Activating Token Index: 971

Text #3

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.9915. Min Act: -0.0000

Data Index: 558622 (C4 (Web Text))

Max Activating Token Index: 59

Text #4

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.9832. Min Act: -0.0000

Data Index: 941482 (C4 (Web Text))

Max Activating Token Index: 562

Text #5

Max Range: 2.5854. Min Range: -2.5854

Max Act: 2.0022. Min Act: -0.0001

Data Index: 964219 (C4 (Web Text))

Max Activating Token Index: 632

Text #6

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.9628. Min Act: -0.0000

Data Index: 448435 (C4 (Web Text))

Max Activating Token Index: 375

Text #7

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.9267. Min Act: -0.0000

Data Index: 1220501 (C4 (Web Text))

Max Activating Token Index: 148

Text #8

Max Range: 2.5854. Min Range: -2.5854

Max Act: 2.2464. Min Act: -0.0000

Data Index: 280700 (C4 (Web Text))

Max Activating Token Index: 875

Text #9

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.5901. Min Act: -0.0000

Data Index: 952943 (C4 (Web Text))

Max Activating Token Index: 823

Text #10

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.8091. Min Act: -0.0000

Data Index: 633869 (C4 (Web Text))

Max Activating Token Index: 287

Text #11

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.7607. Min Act: -0.0001

Data Index: 1146342 (C4 (Web Text))

Max Activating Token Index: 473

Text #12

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.5847. Min Act: -0.0000

Data Index: 1496875 (Python Code)

Max Activating Token Index: 585

Text #13

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.7637. Min Act: -0.0000

Data Index: 21100 (C4 (Web Text))

Max Activating Token Index: 63

Text #14

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.7067. Min Act: -0.0000

Data Index: 1527998 (Python Code)

Max Activating Token Index: 756

Text #15

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.5732. Min Act: -0.0000

Data Index: 1581659 (Python Code)

Max Activating Token Index: 954

Text #16

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.9600. Min Act: -0.0000

Data Index: 1094749 (C4 (Web Text))

Max Activating Token Index: 839

Text #17

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.5771. Min Act: -0.0001

Data Index: 1585841 (Python Code)

Max Activating Token Index: 784

Text #18

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.6738. Min Act: -0.0000

Data Index: 1428750 (Python Code)

Max Activating Token Index: 132

Text #19

Max Range: 2.5854. Min Range: -2.5854

Max Act: 1.5899. Min Act: -0.0000

Data Index: 1585651 (Python Code)

Max Activating Token Index: 823