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

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

Neuron 1057 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0936. Min Act: -0.0002

Data Index: 559464 (C4 (Web Text))

Max Activating Token Index: 47

Text #1

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0699. Min Act: -0.0002

Data Index: 1263495 (C4 (Web Text))

Max Activating Token Index: 21

Text #2

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0682. Min Act: -0.0002

Data Index: 1144754 (C4 (Web Text))

Max Activating Token Index: 214

Text #3

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0681. Min Act: -0.0002

Data Index: 222425 (C4 (Web Text))

Max Activating Token Index: 572

Text #4

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0751. Min Act: -0.0002

Data Index: 1242615 (C4 (Web Text))

Max Activating Token Index: 42

Text #5

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0671. Min Act: -0.0002

Data Index: 867415 (C4 (Web Text))

Max Activating Token Index: 196

Text #6

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0609. Min Act: -0.0002

Data Index: 1110872 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0676. Min Act: -0.0002

Data Index: 1344134 (C4 (Web Text))

Max Activating Token Index: 54

Text #8

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0692. Min Act: -0.0002

Data Index: 1437946 (Python Code)

Max Activating Token Index: 256

Text #9

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0663. Min Act: -0.0002

Data Index: 1151397 (C4 (Web Text))

Max Activating Token Index: 16

Text #10

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0575. Min Act: -0.0002

Data Index: 770207 (C4 (Web Text))

Max Activating Token Index: 976

Text #11

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0506. Min Act: -0.0002

Data Index: 1080241 (C4 (Web Text))

Max Activating Token Index: 755

Text #12

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0644. Min Act: -0.0002

Data Index: 167187 (C4 (Web Text))

Max Activating Token Index: 58

Text #13

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0621. Min Act: -0.0002

Data Index: 1543430 (Python Code)

Max Activating Token Index: 16

Text #14

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0620. Min Act: -0.0002

Data Index: 866269 (C4 (Web Text))

Max Activating Token Index: 105

Text #15

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0596. Min Act: -0.0002

Data Index: 1296721 (C4 (Web Text))

Max Activating Token Index: 379

Text #16

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0559. Min Act: -0.0002

Data Index: 241301 (C4 (Web Text))

Max Activating Token Index: 99

Text #17

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0607. Min Act: -0.0002

Data Index: 1609157 (Python Code)

Max Activating Token Index: 530

Text #18

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0606. Min Act: -0.0002

Data Index: 377602 (C4 (Web Text))

Max Activating Token Index: 36

Text #19

Max Range: 0.0936. Min Range: -0.0936

Max Act: 0.0563. Min Act: -0.0002

Data Index: 1571910 (Python Code)

Max Activating Token Index: 626