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

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

Neuron 1196 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: 0.2886. Min Range: -0.2886

Max Act: 0.2886. Min Act: -0.0000

Data Index: 643180 (C4 (Web Text))

Max Activating Token Index: 155

Text #1

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2855. Min Act: -0.0000

Data Index: 1293603 (C4 (Web Text))

Max Activating Token Index: 22

Text #2

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2266. Min Act: -0.0000

Data Index: 575819 (C4 (Web Text))

Max Activating Token Index: 884

Text #3

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2293. Min Act: -0.0000

Data Index: 1009863 (C4 (Web Text))

Max Activating Token Index: 889

Text #4

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2291. Min Act: -0.0000

Data Index: 803054 (C4 (Web Text))

Max Activating Token Index: 905

Text #5

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2236. Min Act: -0.0000

Data Index: 1132859 (C4 (Web Text))

Max Activating Token Index: 690

Text #6

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2116. Min Act: -0.0000

Data Index: 853079 (C4 (Web Text))

Max Activating Token Index: 60

Text #7

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2114. Min Act: -0.0001

Data Index: 603823 (C4 (Web Text))

Max Activating Token Index: 419

Text #8

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2105. Min Act: -0.0000

Data Index: 207082 (C4 (Web Text))

Max Activating Token Index: 45

Text #9

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2072. Min Act: -0.0000

Data Index: 865180 (C4 (Web Text))

Max Activating Token Index: 375

Text #10

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2114. Min Act: -0.0000

Data Index: 668641 (C4 (Web Text))

Max Activating Token Index: 514

Text #11

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2006. Min Act: -0.0000

Data Index: 1131325 (C4 (Web Text))

Max Activating Token Index: 650

Text #12

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.1977. Min Act: -0.0000

Data Index: 427363 (C4 (Web Text))

Max Activating Token Index: 618

Text #13

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2019. Min Act: -0.0001

Data Index: 352638 (C4 (Web Text))

Max Activating Token Index: 269

Text #14

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.1989. Min Act: -0.0001

Data Index: 537818 (C4 (Web Text))

Max Activating Token Index: 244

Text #15

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.1980. Min Act: -0.0001

Data Index: 867856 (C4 (Web Text))

Max Activating Token Index: 9

Text #16

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2040. Min Act: -0.0000

Data Index: 1001964 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2020. Min Act: -0.0000

Data Index: 672385 (C4 (Web Text))

Max Activating Token Index: 792

Text #18

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.2022. Min Act: -0.0000

Data Index: 496002 (C4 (Web Text))

Max Activating Token Index: 710

Text #19

Max Range: 0.2886. Min Range: -0.2886

Max Act: 0.1965. Min Act: -0.0001

Data Index: 68239 (C4 (Web Text))

Max Activating Token Index: 974