Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3186 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.6122. Min Act: -0.0000

Data Index: 1431989 (Python Code)

Max Activating Token Index: 651

Text #1

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.6515. Min Act: -0.0000

Data Index: 1588965 (Python Code)

Max Activating Token Index: 993

Text #2

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.6358. Min Act: -0.0000

Data Index: 1611542 (Python Code)

Max Activating Token Index: 334

Text #3

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.5351. Min Act: 0.0001

Data Index: 1589388 (Python Code)

Max Activating Token Index: 283

Text #4

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.5367. Min Act: -0.0000

Data Index: 1353351 (C4 (Web Text))

Max Activating Token Index: 751

Text #5

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4832. Min Act: -0.0000

Data Index: 516396 (C4 (Web Text))

Max Activating Token Index: 104

Text #6

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.5049. Min Act: -0.0000

Data Index: 64036 (C4 (Web Text))

Max Activating Token Index: 742

Text #7

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4857. Min Act: -0.0000

Data Index: 15072 (C4 (Web Text))

Max Activating Token Index: 833

Text #8

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4760. Min Act: -0.0000

Data Index: 20692 (C4 (Web Text))

Max Activating Token Index: 583

Text #9

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4719. Min Act: -0.0000

Data Index: 27492 (C4 (Web Text))

Max Activating Token Index: 908

Text #10

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4405. Min Act: -0.0000

Data Index: 341128 (C4 (Web Text))

Max Activating Token Index: 782

Text #11

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4284. Min Act: -0.0000

Data Index: 1413236 (Python Code)

Max Activating Token Index: 944

Text #12

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4451. Min Act: -0.0000

Data Index: 1403465 (Python Code)

Max Activating Token Index: 1014

Text #13

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4364. Min Act: 0.0000

Data Index: 631358 (C4 (Web Text))

Max Activating Token Index: 460

Text #14

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4482. Min Act: -0.0000

Data Index: 133790 (C4 (Web Text))

Max Activating Token Index: 57

Text #15

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4325. Min Act: -0.0000

Data Index: 952224 (C4 (Web Text))

Max Activating Token Index: 979

Text #16

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4123. Min Act: -0.0000

Data Index: 1637166 (Python Code)

Max Activating Token Index: 683

Text #17

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4346. Min Act: -0.0000

Data Index: 832123 (C4 (Web Text))

Max Activating Token Index: 283

Text #18

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4159. Min Act: -0.0000

Data Index: 326516 (C4 (Web Text))

Max Activating Token Index: 808

Text #19

Max Range: 0.6515. Min Range: -0.6515

Max Act: 0.4166. Min Act: -0.0000

Data Index: 1608238 (Python Code)

Max Activating Token Index: 497