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

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

Neuron 3540 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.2613. Min Act: -0.0001

Data Index: 145993 (C4 (Web Text))

Max Activating Token Index: 304

Text #1

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.1114. Min Act: -0.0001

Data Index: 239282 (C4 (Web Text))

Max Activating Token Index: 738

Text #2

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.1360. Min Act: -0.0001

Data Index: 1568491 (Python Code)

Max Activating Token Index: 865

Text #3

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.1648. Min Act: -0.0001

Data Index: 1496161 (Python Code)

Max Activating Token Index: 484

Text #4

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.0269. Min Act: -0.0001

Data Index: 721347 (C4 (Web Text))

Max Activating Token Index: 554

Text #5

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.0684. Min Act: -0.0001

Data Index: 1305700 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.1037. Min Act: -0.0001

Data Index: 797932 (C4 (Web Text))

Max Activating Token Index: 727

Text #7

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9910. Min Act: -0.0001

Data Index: 1263090 (C4 (Web Text))

Max Activating Token Index: 448

Text #8

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.0762. Min Act: -0.0001

Data Index: 1619715 (Python Code)

Max Activating Token Index: 989

Text #9

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9462. Min Act: -0.0001

Data Index: 1095937 (C4 (Web Text))

Max Activating Token Index: 796

Text #10

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9140. Min Act: -0.0001

Data Index: 1482044 (Python Code)

Max Activating Token Index: 681

Text #11

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9919. Min Act: -0.0001

Data Index: 675872 (C4 (Web Text))

Max Activating Token Index: 709

Text #12

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.0047. Min Act: -0.0001

Data Index: 1181177 (C4 (Web Text))

Max Activating Token Index: 541

Text #13

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.8904. Min Act: -0.0001

Data Index: 1553460 (Python Code)

Max Activating Token Index: 822

Text #14

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9321. Min Act: -0.0001

Data Index: 1374431 (Python Code)

Max Activating Token Index: 461

Text #15

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9257. Min Act: -0.0001

Data Index: 1631140 (Python Code)

Max Activating Token Index: 162

Text #16

Max Range: 1.2613. Min Range: -1.2613

Max Act: 1.0082. Min Act: -0.0001

Data Index: 321136 (C4 (Web Text))

Max Activating Token Index: 751

Text #17

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9513. Min Act: -0.0001

Data Index: 1635155 (Python Code)

Max Activating Token Index: 945

Text #18

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.8934. Min Act: -0.0001

Data Index: 1380670 (Python Code)

Max Activating Token Index: 988

Text #19

Max Range: 1.2613. Min Range: -1.2613

Max Act: 0.9204. Min Act: -0.0001

Data Index: 667763 (C4 (Web Text))

Max Activating Token Index: 305