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

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

Neuron 603 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: 0.1306. Min Range: -0.1306

Max Act: 0.1306. Min Act: -0.0002

Data Index: 314789 (C4 (Web Text))

Max Activating Token Index: 111

Text #1

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1185. Min Act: -0.0001

Data Index: 108469 (C4 (Web Text))

Max Activating Token Index: 638

Text #2

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1210. Min Act: -0.0002

Data Index: 1187021 (C4 (Web Text))

Max Activating Token Index: 901

Text #3

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1046. Min Act: -0.0002

Data Index: 1160445 (C4 (Web Text))

Max Activating Token Index: 461

Text #4

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0995. Min Act: -0.0001

Data Index: 610325 (C4 (Web Text))

Max Activating Token Index: 470

Text #5

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1031. Min Act: -0.0002

Data Index: 246198 (C4 (Web Text))

Max Activating Token Index: 131

Text #6

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0982. Min Act: -0.0001

Data Index: 340176 (C4 (Web Text))

Max Activating Token Index: 989

Text #7

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1003. Min Act: -0.0002

Data Index: 1080908 (C4 (Web Text))

Max Activating Token Index: 712

Text #8

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0989. Min Act: -0.0001

Data Index: 1635853 (Python Code)

Max Activating Token Index: 395

Text #9

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0963. Min Act: -0.0002

Data Index: 199618 (C4 (Web Text))

Max Activating Token Index: 930

Text #10

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1168. Min Act: -0.0002

Data Index: 966295 (C4 (Web Text))

Max Activating Token Index: 167

Text #11

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0944. Min Act: -0.0002

Data Index: 1075796 (C4 (Web Text))

Max Activating Token Index: 279

Text #12

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.1037. Min Act: -0.0002

Data Index: 198261 (C4 (Web Text))

Max Activating Token Index: 290

Text #13

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0994. Min Act: -0.0002

Data Index: 384392 (C4 (Web Text))

Max Activating Token Index: 246

Text #14

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0953. Min Act: -0.0002

Data Index: 716047 (C4 (Web Text))

Max Activating Token Index: 849

Text #15

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0923. Min Act: -0.0002

Data Index: 1104122 (C4 (Web Text))

Max Activating Token Index: 177

Text #16

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0890. Min Act: -0.0002

Data Index: 103527 (C4 (Web Text))

Max Activating Token Index: 453

Text #17

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0846. Min Act: -0.0002

Data Index: 329545 (C4 (Web Text))

Max Activating Token Index: 660

Text #18

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0901. Min Act: -0.0002

Data Index: 1304652 (C4 (Web Text))

Max Activating Token Index: 406

Text #19

Max Range: 0.1306. Min Range: -0.1306

Max Act: 0.0935. Min Act: -0.0002

Data Index: 254293 (C4 (Web Text))

Max Activating Token Index: 419