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

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

Neuron 3356 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: 0.1162. Min Range: -0.1162

Max Act: 0.1162. Min Act: -0.0001

Data Index: 690520 (C4 (Web Text))

Max Activating Token Index: 822

Text #1

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1145. Min Act: -0.0001

Data Index: 767654 (C4 (Web Text))

Max Activating Token Index: 659

Text #2

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1100. Min Act: -0.0001

Data Index: 1394282 (Python Code)

Max Activating Token Index: 567

Text #3

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1039. Min Act: -0.0001

Data Index: 1118943 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1094. Min Act: -0.0001

Data Index: 339674 (C4 (Web Text))

Max Activating Token Index: 461

Text #5

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1063. Min Act: -0.0001

Data Index: 1470391 (Python Code)

Max Activating Token Index: 56

Text #6

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1012. Min Act: -0.0001

Data Index: 685808 (C4 (Web Text))

Max Activating Token Index: 922

Text #7

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0989. Min Act: -0.0001

Data Index: 638730 (C4 (Web Text))

Max Activating Token Index: 41

Text #8

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.1014. Min Act: -0.0001

Data Index: 698260 (C4 (Web Text))

Max Activating Token Index: 685

Text #9

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0919. Min Act: -0.0001

Data Index: 536804 (C4 (Web Text))

Max Activating Token Index: 50

Text #10

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0948. Min Act: -0.0000

Data Index: 1625105 (Python Code)

Max Activating Token Index: 65

Text #11

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0880. Min Act: -0.0001

Data Index: 1204553 (C4 (Web Text))

Max Activating Token Index: 841

Text #12

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0942. Min Act: -0.0001

Data Index: 242931 (C4 (Web Text))

Max Activating Token Index: 77

Text #13

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0935. Min Act: -0.0001

Data Index: 1455392 (Python Code)

Max Activating Token Index: 744

Text #14

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0892. Min Act: -0.0001

Data Index: 1361831 (Python Code)

Max Activating Token Index: 289

Text #15

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0887. Min Act: -0.0001

Data Index: 1267612 (C4 (Web Text))

Max Activating Token Index: 40

Text #16

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0879. Min Act: -0.0001

Data Index: 1483147 (Python Code)

Max Activating Token Index: 175

Text #17

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0941. Min Act: -0.0001

Data Index: 630852 (C4 (Web Text))

Max Activating Token Index: 972

Text #18

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0878. Min Act: -0.0001

Data Index: 1097247 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 0.1162. Min Range: -0.1162

Max Act: 0.0913. Min Act: -0.0001

Data Index: 830177 (C4 (Web Text))

Max Activating Token Index: 42