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

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

Neuron 2231 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.5948. Min Act: -0.0000

Data Index: 1497265 (Python Code)

Max Activating Token Index: 260

Text #1

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4449. Min Act: -0.0000

Data Index: 1010939 (C4 (Web Text))

Max Activating Token Index: 271

Text #2

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.3715. Min Act: -0.0000

Data Index: 1324891 (C4 (Web Text))

Max Activating Token Index: 265

Text #3

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4461. Min Act: -0.0000

Data Index: 1540549 (Python Code)

Max Activating Token Index: 921

Text #4

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4999. Min Act: -0.0000

Data Index: 440398 (C4 (Web Text))

Max Activating Token Index: 601

Text #5

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4256. Min Act: -0.0000

Data Index: 1124998 (C4 (Web Text))

Max Activating Token Index: 395

Text #6

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4130. Min Act: -0.0000

Data Index: 280833 (C4 (Web Text))

Max Activating Token Index: 146

Text #7

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.3976. Min Act: -0.0000

Data Index: 1476879 (Python Code)

Max Activating Token Index: 19

Text #8

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.3873. Min Act: -0.0000

Data Index: 1100894 (C4 (Web Text))

Max Activating Token Index: 359

Text #9

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2835. Min Act: -0.0000

Data Index: 1095904 (C4 (Web Text))

Max Activating Token Index: 298

Text #10

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.3680. Min Act: -0.0000

Data Index: 1275941 (C4 (Web Text))

Max Activating Token Index: 305

Text #11

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.3274. Min Act: -0.0000

Data Index: 1372638 (Python Code)

Max Activating Token Index: 161

Text #12

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.4082. Min Act: -0.0000

Data Index: 1234366 (C4 (Web Text))

Max Activating Token Index: 616

Text #13

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2629. Min Act: -0.0000

Data Index: 1198970 (C4 (Web Text))

Max Activating Token Index: 214

Text #14

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2518. Min Act: -0.0000

Data Index: 1438920 (Python Code)

Max Activating Token Index: 361

Text #15

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.1441. Min Act: -0.0000

Data Index: 254013 (C4 (Web Text))

Max Activating Token Index: 185

Text #16

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2129. Min Act: -0.0000

Data Index: 1071821 (C4 (Web Text))

Max Activating Token Index: 620

Text #17

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2873. Min Act: -0.0000

Data Index: 940206 (C4 (Web Text))

Max Activating Token Index: 875

Text #18

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2279. Min Act: -0.0000

Data Index: 1651649 (Python Code)

Max Activating Token Index: 722

Text #19

Max Range: 1.5948. Min Range: -1.5948

Max Act: 1.2315. Min Act: -0.0000

Data Index: 667016 (C4 (Web Text))

Max Activating Token Index: 400