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

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

Neuron 3714 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.7702. Min Act: -0.0001

Data Index: 1210297 (C4 (Web Text))

Max Activating Token Index: 793

Text #1

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.8515. Min Act: -0.0001

Data Index: 357042 (C4 (Web Text))

Max Activating Token Index: 984

Text #2

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5859. Min Act: -0.0001

Data Index: 708227 (C4 (Web Text))

Max Activating Token Index: 881

Text #3

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.6081. Min Act: -0.0001

Data Index: 877768 (C4 (Web Text))

Max Activating Token Index: 371

Text #4

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5919. Min Act: -0.0001

Data Index: 738281 (C4 (Web Text))

Max Activating Token Index: 772

Text #5

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.7129. Min Act: -0.0001

Data Index: 1377577 (Python Code)

Max Activating Token Index: 662

Text #6

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.4792. Min Act: -0.0001

Data Index: 717417 (C4 (Web Text))

Max Activating Token Index: 250

Text #7

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5183. Min Act: -0.0001

Data Index: 981552 (C4 (Web Text))

Max Activating Token Index: 801

Text #8

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5809. Min Act: -0.0001

Data Index: 1278447 (C4 (Web Text))

Max Activating Token Index: 860

Text #9

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5275. Min Act: -0.0001

Data Index: 1554570 (Python Code)

Max Activating Token Index: 966

Text #10

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5287. Min Act: -0.0001

Data Index: 1391168 (Python Code)

Max Activating Token Index: 816

Text #11

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.4857. Min Act: -0.0001

Data Index: 1218161 (C4 (Web Text))

Max Activating Token Index: 281

Text #12

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.6196. Min Act: -0.0001

Data Index: 32694 (C4 (Web Text))

Max Activating Token Index: 307

Text #13

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.6881. Min Act: -0.0001

Data Index: 1554382 (Python Code)

Max Activating Token Index: 812

Text #14

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.3827. Min Act: -0.0001

Data Index: 526526 (C4 (Web Text))

Max Activating Token Index: 336

Text #15

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.5319. Min Act: -0.0001

Data Index: 1518841 (Python Code)

Max Activating Token Index: 666

Text #16

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.4877. Min Act: -0.0001

Data Index: 1399677 (Python Code)

Max Activating Token Index: 623

Text #17

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.6505. Min Act: -0.0001

Data Index: 455479 (C4 (Web Text))

Max Activating Token Index: 581

Text #18

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.4801. Min Act: -0.0001

Data Index: 967474 (C4 (Web Text))

Max Activating Token Index: 981

Text #19

Max Range: 1.8515. Min Range: -1.8515

Max Act: 1.4741. Min Act: -0.0001

Data Index: 1613863 (Python Code)

Max Activating Token Index: 549