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

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

Neuron 3897 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1700. Min Act: -0.0001

Data Index: 20730 (C4 (Web Text))

Max Activating Token Index: 905

Text #1

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1605. Min Act: -0.0001

Data Index: 1635973 (Python Code)

Max Activating Token Index: 292

Text #2

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1667. Min Act: -0.0001

Data Index: 1381563 (Python Code)

Max Activating Token Index: 80

Text #3

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1605. Min Act: -0.0000

Data Index: 1276877 (C4 (Web Text))

Max Activating Token Index: 823

Text #4

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1721. Min Act: -0.0000

Data Index: 703720 (C4 (Web Text))

Max Activating Token Index: 665

Text #5

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1505. Min Act: -0.0001

Data Index: 1535710 (Python Code)

Max Activating Token Index: 375

Text #6

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1578. Min Act: -0.0001

Data Index: 828973 (C4 (Web Text))

Max Activating Token Index: 237

Text #7

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1314. Min Act: -0.0000

Data Index: 1030190 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1502. Min Act: -0.0001

Data Index: 828627 (C4 (Web Text))

Max Activating Token Index: 350

Text #9

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1508. Min Act: -0.0000

Data Index: 1012371 (C4 (Web Text))

Max Activating Token Index: 696

Text #10

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1487. Min Act: -0.0000

Data Index: 1002421 (C4 (Web Text))

Max Activating Token Index: 715

Text #11

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1460. Min Act: -0.0001

Data Index: 1388550 (Python Code)

Max Activating Token Index: 615

Text #12

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1402. Min Act: -0.0001

Data Index: 1402619 (Python Code)

Max Activating Token Index: 486

Text #13

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1440. Min Act: -0.0001

Data Index: 608207 (C4 (Web Text))

Max Activating Token Index: 544

Text #14

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1438. Min Act: -0.0001

Data Index: 1511896 (Python Code)

Max Activating Token Index: 549

Text #15

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1392. Min Act: -0.0001

Data Index: 954315 (C4 (Web Text))

Max Activating Token Index: 748

Text #16

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1319. Min Act: -0.0001

Data Index: 932451 (C4 (Web Text))

Max Activating Token Index: 602

Text #17

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1437. Min Act: -0.0001

Data Index: 1363851 (Python Code)

Max Activating Token Index: 352

Text #18

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1316. Min Act: -0.0001

Data Index: 1383518 (Python Code)

Max Activating Token Index: 488

Text #19

Max Range: 0.1721. Min Range: -0.1721

Max Act: 0.1359. Min Act: -0.0001

Data Index: 1525728 (Python Code)

Max Activating Token Index: 692