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

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

Neuron 1590 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.7208. Min Act: -0.0001

Data Index: 214503 (C4 (Web Text))

Max Activating Token Index: 551

Text #1

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.7598. Min Act: -0.0002

Data Index: 956028 (C4 (Web Text))

Max Activating Token Index: 448

Text #2

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6560. Min Act: -0.0001

Data Index: 418474 (C4 (Web Text))

Max Activating Token Index: 689

Text #3

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6767. Min Act: -0.0001

Data Index: 312267 (C4 (Web Text))

Max Activating Token Index: 12

Text #4

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5922. Min Act: -0.0001

Data Index: 1175626 (C4 (Web Text))

Max Activating Token Index: 889

Text #5

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.7098. Min Act: -0.0002

Data Index: 984519 (C4 (Web Text))

Max Activating Token Index: 416

Text #6

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6449. Min Act: -0.0002

Data Index: 652340 (C4 (Web Text))

Max Activating Token Index: 608

Text #7

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6466. Min Act: -0.0002

Data Index: 28999 (C4 (Web Text))

Max Activating Token Index: 1002

Text #8

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6549. Min Act: -0.0001

Data Index: 821132 (C4 (Web Text))

Max Activating Token Index: 111

Text #9

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6958. Min Act: -0.0001

Data Index: 406685 (C4 (Web Text))

Max Activating Token Index: 815

Text #10

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5981. Min Act: -0.0002

Data Index: 355966 (C4 (Web Text))

Max Activating Token Index: 340

Text #11

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6762. Min Act: -0.0002

Data Index: 1292898 (C4 (Web Text))

Max Activating Token Index: 931

Text #12

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.7221. Min Act: -0.0002

Data Index: 968964 (C4 (Web Text))

Max Activating Token Index: 901

Text #13

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6181. Min Act: -0.0001

Data Index: 1415313 (Python Code)

Max Activating Token Index: 43

Text #14

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5974. Min Act: -0.0002

Data Index: 227790 (C4 (Web Text))

Max Activating Token Index: 904

Text #15

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5997. Min Act: -0.0001

Data Index: 653644 (C4 (Web Text))

Max Activating Token Index: 823

Text #16

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6630. Min Act: -0.0002

Data Index: 420935 (C4 (Web Text))

Max Activating Token Index: 542

Text #17

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5156. Min Act: -0.0002

Data Index: 60975 (C4 (Web Text))

Max Activating Token Index: 657

Text #18

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.6024. Min Act: -0.0001

Data Index: 1655915 (Python Code)

Max Activating Token Index: 38

Text #19

Max Range: 0.7598. Min Range: -0.7598

Max Act: 0.5624. Min Act: -0.0001

Data Index: 913184 (C4 (Web Text))

Max Activating Token Index: 419