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

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

Neuron 3877 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6008. Min Act: -0.0001

Data Index: 1598191 (Python Code)

Max Activating Token Index: 671

Text #1

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6657. Min Act: -0.0001

Data Index: 1579748 (Python Code)

Max Activating Token Index: 921

Text #2

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6373. Min Act: -0.0001

Data Index: 1512532 (Python Code)

Max Activating Token Index: 986

Text #3

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6074. Min Act: -0.0001

Data Index: 1508818 (Python Code)

Max Activating Token Index: 140

Text #4

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6261. Min Act: -0.0001

Data Index: 1590177 (Python Code)

Max Activating Token Index: 954

Text #5

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5871. Min Act: -0.0000

Data Index: 1549609 (Python Code)

Max Activating Token Index: 732

Text #6

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6336. Min Act: -0.0000

Data Index: 1646028 (Python Code)

Max Activating Token Index: 848

Text #7

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6050. Min Act: -0.0000

Data Index: 63286 (C4 (Web Text))

Max Activating Token Index: 46

Text #8

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6076. Min Act: -0.0001

Data Index: 1592296 (Python Code)

Max Activating Token Index: 714

Text #9

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6034. Min Act: -0.0000

Data Index: 359122 (C4 (Web Text))

Max Activating Token Index: 452

Text #10

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5852. Min Act: -0.0001

Data Index: 1532226 (Python Code)

Max Activating Token Index: 491

Text #11

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5768. Min Act: -0.0001

Data Index: 1603380 (Python Code)

Max Activating Token Index: 371

Text #12

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5330. Min Act: -0.0001

Data Index: 1616823 (Python Code)

Max Activating Token Index: 188

Text #13

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5556. Min Act: -0.0001

Data Index: 1565720 (Python Code)

Max Activating Token Index: 348

Text #14

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5710. Min Act: -0.0001

Data Index: 1411523 (Python Code)

Max Activating Token Index: 146

Text #15

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5667. Min Act: -0.0001

Data Index: 321950 (C4 (Web Text))

Max Activating Token Index: 659

Text #16

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5947. Min Act: -0.0001

Data Index: 1647302 (Python Code)

Max Activating Token Index: 600

Text #17

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5436. Min Act: -0.0001

Data Index: 1629539 (Python Code)

Max Activating Token Index: 158

Text #18

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.6080. Min Act: -0.0001

Data Index: 1380295 (Python Code)

Max Activating Token Index: 147

Text #19

Max Range: 0.6657. Min Range: -0.6657

Max Act: 0.5125. Min Act: -0.0000

Data Index: 654829 (C4 (Web Text))

Max Activating Token Index: 915