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

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

Neuron 1040 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4539. Min Act: -0.0001

Data Index: 465106 (C4 (Web Text))

Max Activating Token Index: 1002

Text #1

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4659. Min Act: -0.0001

Data Index: 151478 (C4 (Web Text))

Max Activating Token Index: 990

Text #2

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4284. Min Act: -0.0001

Data Index: 1010855 (C4 (Web Text))

Max Activating Token Index: 310

Text #3

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4327. Min Act: -0.0001

Data Index: 344072 (C4 (Web Text))

Max Activating Token Index: 104

Text #4

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4003. Min Act: -0.0001

Data Index: 715948 (C4 (Web Text))

Max Activating Token Index: 544

Text #5

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4276. Min Act: -0.0001

Data Index: 1096426 (C4 (Web Text))

Max Activating Token Index: 1010

Text #6

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4394. Min Act: -0.0001

Data Index: 363188 (C4 (Web Text))

Max Activating Token Index: 61

Text #7

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4059. Min Act: -0.0001

Data Index: 534418 (C4 (Web Text))

Max Activating Token Index: 127

Text #8

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4184. Min Act: -0.0001

Data Index: 730864 (C4 (Web Text))

Max Activating Token Index: 986

Text #9

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4201. Min Act: -0.0001

Data Index: 476633 (C4 (Web Text))

Max Activating Token Index: 92

Text #10

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.3792. Min Act: -0.0001

Data Index: 516913 (C4 (Web Text))

Max Activating Token Index: 846

Text #11

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4037. Min Act: -0.0001

Data Index: 265326 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4013. Min Act: -0.0001

Data Index: 479343 (C4 (Web Text))

Max Activating Token Index: 388

Text #13

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4031. Min Act: -0.0001

Data Index: 589699 (C4 (Web Text))

Max Activating Token Index: 438

Text #14

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4012. Min Act: -0.0001

Data Index: 169342 (C4 (Web Text))

Max Activating Token Index: 922

Text #15

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4281. Min Act: -0.0001

Data Index: 945292 (C4 (Web Text))

Max Activating Token Index: 216

Text #16

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4142. Min Act: -0.0001

Data Index: 768863 (C4 (Web Text))

Max Activating Token Index: 975

Text #17

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4119. Min Act: -0.0001

Data Index: 421804 (C4 (Web Text))

Max Activating Token Index: 973

Text #18

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.3843. Min Act: -0.0001

Data Index: 1120770 (C4 (Web Text))

Max Activating Token Index: 732

Text #19

Max Range: 0.4659. Min Range: -0.4659

Max Act: 0.4087. Min Act: -0.0001

Data Index: 193168 (C4 (Web Text))

Max Activating Token Index: 102