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

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

Neuron 123 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1432. Min Act: -0.0001

Data Index: 405701 (C4 (Web Text))

Max Activating Token Index: 75

Text #1

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1315. Min Act: -0.0001

Data Index: 971938 (C4 (Web Text))

Max Activating Token Index: 567

Text #2

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1158. Min Act: -0.0001

Data Index: 686998 (C4 (Web Text))

Max Activating Token Index: 347

Text #3

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1195. Min Act: -0.0000

Data Index: 467823 (C4 (Web Text))

Max Activating Token Index: 704

Text #4

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1128. Min Act: -0.0000

Data Index: 852745 (C4 (Web Text))

Max Activating Token Index: 835

Text #5

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1105. Min Act: -0.0001

Data Index: 909121 (C4 (Web Text))

Max Activating Token Index: 295

Text #6

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1068. Min Act: -0.0000

Data Index: 527524 (C4 (Web Text))

Max Activating Token Index: 272

Text #7

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1043. Min Act: -0.0000

Data Index: 148777 (C4 (Web Text))

Max Activating Token Index: 147

Text #8

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0969. Min Act: -0.0001

Data Index: 1333381 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1031. Min Act: -0.0000

Data Index: 82121 (C4 (Web Text))

Max Activating Token Index: 632

Text #10

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0987. Min Act: -0.0000

Data Index: 611185 (C4 (Web Text))

Max Activating Token Index: 496

Text #11

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.1043. Min Act: -0.0001

Data Index: 684902 (C4 (Web Text))

Max Activating Token Index: 408

Text #12

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0955. Min Act: -0.0000

Data Index: 747621 (C4 (Web Text))

Max Activating Token Index: 459

Text #13

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0964. Min Act: -0.0000

Data Index: 196708 (C4 (Web Text))

Max Activating Token Index: 681

Text #14

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0927. Min Act: -0.0000

Data Index: 164477 (C4 (Web Text))

Max Activating Token Index: 404

Text #15

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0956. Min Act: -0.0001

Data Index: 725546 (C4 (Web Text))

Max Activating Token Index: 860

Text #16

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0944. Min Act: -0.0001

Data Index: 1463892 (Python Code)

Max Activating Token Index: 732

Text #17

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0947. Min Act: -0.0000

Data Index: 902346 (C4 (Web Text))

Max Activating Token Index: 297

Text #18

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0975. Min Act: -0.0000

Data Index: 923202 (C4 (Web Text))

Max Activating Token Index: 342

Text #19

Max Range: 0.1432. Min Range: -0.1432

Max Act: 0.0929. Min Act: -0.0000

Data Index: 781833 (C4 (Web Text))

Max Activating Token Index: 261