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

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

Neuron 976 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8295. Min Range: -3.8295

Max Act: 3.7388. Min Act: -0.0001

Data Index: 1575661 (Python Code)

Max Activating Token Index: 595

Text #1

Max Range: 3.8295. Min Range: -3.8295

Max Act: 3.8295. Min Act: -0.0001

Data Index: 1459310 (Python Code)

Max Activating Token Index: 520

Text #2

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.9991. Min Act: -0.0001

Data Index: 1633138 (Python Code)

Max Activating Token Index: 301

Text #3

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.7083. Min Act: -0.0001

Data Index: 1519469 (Python Code)

Max Activating Token Index: 391

Text #4

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.6726. Min Act: -0.0001

Data Index: 1085806 (C4 (Web Text))

Max Activating Token Index: 962

Text #5

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.8834. Min Act: -0.0001

Data Index: 310553 (C4 (Web Text))

Max Activating Token Index: 40

Text #6

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.5811. Min Act: -0.0001

Data Index: 1563964 (Python Code)

Max Activating Token Index: 525

Text #7

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.3128. Min Act: -0.0001

Data Index: 1438948 (Python Code)

Max Activating Token Index: 367

Text #8

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.3289. Min Act: -0.0001

Data Index: 103958 (C4 (Web Text))

Max Activating Token Index: 493

Text #9

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.2799. Min Act: -0.0000

Data Index: 11367 (C4 (Web Text))

Max Activating Token Index: 595

Text #10

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.2594. Min Act: -0.0001

Data Index: 348117 (C4 (Web Text))

Max Activating Token Index: 269

Text #11

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.2717. Min Act: -0.0001

Data Index: 860651 (C4 (Web Text))

Max Activating Token Index: 548

Text #12

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.1383. Min Act: -0.0001

Data Index: 586625 (C4 (Web Text))

Max Activating Token Index: 949

Text #13

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.3726. Min Act: -0.0001

Data Index: 1628283 (Python Code)

Max Activating Token Index: 401

Text #14

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.2144. Min Act: -0.0001

Data Index: 591738 (C4 (Web Text))

Max Activating Token Index: 479

Text #15

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.1253. Min Act: -0.0001

Data Index: 940176 (C4 (Web Text))

Max Activating Token Index: 245

Text #16

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.1000. Min Act: -0.0001

Data Index: 434232 (C4 (Web Text))

Max Activating Token Index: 47

Text #17

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.2184. Min Act: -0.0000

Data Index: 854579 (C4 (Web Text))

Max Activating Token Index: 718

Text #18

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.3275. Min Act: -0.0000

Data Index: 332607 (C4 (Web Text))

Max Activating Token Index: 656

Text #19

Max Range: 3.8295. Min Range: -3.8295

Max Act: 2.0845. Min Act: -0.0000

Data Index: 1103460 (C4 (Web Text))

Max Activating Token Index: 112