Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1236 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.7446. Min Act: -0.0001

Data Index: 1463806 (Python Code)

Max Activating Token Index: 12

Text #1

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.7242. Min Act: -0.0001

Data Index: 1436941 (Python Code)

Max Activating Token Index: 1022

Text #2

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.7136. Min Act: -0.0001

Data Index: 1339910 (C4 (Web Text))

Max Activating Token Index: 712

Text #3

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.6317. Min Act: -0.0001

Data Index: 992915 (C4 (Web Text))

Max Activating Token Index: 528

Text #4

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.6089. Min Act: -0.0001

Data Index: 1425377 (Python Code)

Max Activating Token Index: 24

Text #5

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5954. Min Act: -0.0001

Data Index: 462427 (C4 (Web Text))

Max Activating Token Index: 804

Text #6

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5618. Min Act: -0.0001

Data Index: 1589691 (Python Code)

Max Activating Token Index: 253

Text #7

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.6175. Min Act: -0.0001

Data Index: 172709 (C4 (Web Text))

Max Activating Token Index: 20

Text #8

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5813. Min Act: -0.0001

Data Index: 452946 (C4 (Web Text))

Max Activating Token Index: 22

Text #9

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5941. Min Act: -0.0001

Data Index: 1508041 (Python Code)

Max Activating Token Index: 1010

Text #10

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5932. Min Act: -0.0001

Data Index: 433569 (C4 (Web Text))

Max Activating Token Index: 1019

Text #11

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5806. Min Act: -0.0001

Data Index: 60062 (C4 (Web Text))

Max Activating Token Index: 874

Text #12

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5870. Min Act: -0.0001

Data Index: 1534788 (Python Code)

Max Activating Token Index: 46

Text #13

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5863. Min Act: -0.0001

Data Index: 45346 (C4 (Web Text))

Max Activating Token Index: 893

Text #14

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5282. Min Act: -0.0001

Data Index: 1019724 (C4 (Web Text))

Max Activating Token Index: 926

Text #15

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5773. Min Act: -0.0001

Data Index: 1136154 (C4 (Web Text))

Max Activating Token Index: 712

Text #16

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5734. Min Act: -0.0001

Data Index: 785579 (C4 (Web Text))

Max Activating Token Index: 979

Text #17

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5600. Min Act: -0.0001

Data Index: 1459955 (Python Code)

Max Activating Token Index: 23

Text #18

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5758. Min Act: -0.0001

Data Index: 1481577 (Python Code)

Max Activating Token Index: 202

Text #19

Max Range: 0.7446. Min Range: -0.7446

Max Act: 0.5773. Min Act: -0.0001

Data Index: 1351197 (C4 (Web Text))

Max Activating Token Index: 568