Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3967 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.5155. Min Act: -0.0000

Data Index: 751829 (C4 (Web Text))

Max Activating Token Index: 44

Text #1

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4901. Min Act: -0.0000

Data Index: 203656 (C4 (Web Text))

Max Activating Token Index: 985

Text #2

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4412. Min Act: -0.0000

Data Index: 1237512 (C4 (Web Text))

Max Activating Token Index: 678

Text #3

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4932. Min Act: -0.0001

Data Index: 371234 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4302. Min Act: -0.0000

Data Index: 1189460 (C4 (Web Text))

Max Activating Token Index: 614

Text #5

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4316. Min Act: -0.0001

Data Index: 102203 (C4 (Web Text))

Max Activating Token Index: 911

Text #6

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4255. Min Act: -0.0000

Data Index: 361796 (C4 (Web Text))

Max Activating Token Index: 896

Text #7

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4131. Min Act: -0.0000

Data Index: 754349 (C4 (Web Text))

Max Activating Token Index: 677

Text #8

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4187. Min Act: -0.0000

Data Index: 63494 (C4 (Web Text))

Max Activating Token Index: 742

Text #9

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4324. Min Act: -0.0000

Data Index: 793040 (C4 (Web Text))

Max Activating Token Index: 711

Text #10

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4309. Min Act: -0.0001

Data Index: 257599 (C4 (Web Text))

Max Activating Token Index: 784

Text #11

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4198. Min Act: -0.0000

Data Index: 210790 (C4 (Web Text))

Max Activating Token Index: 23

Text #12

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4304. Min Act: -0.0001

Data Index: 955853 (C4 (Web Text))

Max Activating Token Index: 409

Text #13

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.4029. Min Act: -0.0001

Data Index: 799392 (C4 (Web Text))

Max Activating Token Index: 33

Text #14

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3843. Min Act: -0.0000

Data Index: 829352 (C4 (Web Text))

Max Activating Token Index: 354

Text #15

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3771. Min Act: -0.0000

Data Index: 449754 (C4 (Web Text))

Max Activating Token Index: 469

Text #16

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3957. Min Act: -0.0001

Data Index: 1429068 (Python Code)

Max Activating Token Index: 1011

Text #17

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3840. Min Act: -0.0001

Data Index: 1180514 (C4 (Web Text))

Max Activating Token Index: 59

Text #18

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3832. Min Act: -0.0000

Data Index: 1409785 (Python Code)

Max Activating Token Index: 66

Text #19

Max Range: 0.5155. Min Range: -0.5155

Max Act: 0.3798. Min Act: -0.0000

Data Index: 436678 (C4 (Web Text))

Max Activating Token Index: 967