Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1670 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1532. Min Act: -0.0001

Data Index: 1057113 (C4 (Web Text))

Max Activating Token Index: 555

Text #1

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1556. Min Act: -0.0001

Data Index: 85922 (C4 (Web Text))

Max Activating Token Index: 443

Text #2

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1651. Min Act: -0.0001

Data Index: 429662 (C4 (Web Text))

Max Activating Token Index: 116

Text #3

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1506. Min Act: -0.0001

Data Index: 1082369 (C4 (Web Text))

Max Activating Token Index: 450

Text #4

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1670. Min Act: -0.0001

Data Index: 189401 (C4 (Web Text))

Max Activating Token Index: 479

Text #5

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1400. Min Act: -0.0001

Data Index: 1147296 (C4 (Web Text))

Max Activating Token Index: 555

Text #6

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1456. Min Act: -0.0001

Data Index: 1068627 (C4 (Web Text))

Max Activating Token Index: 357

Text #7

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1604. Min Act: -0.0001

Data Index: 1015545 (C4 (Web Text))

Max Activating Token Index: 67

Text #8

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1663. Min Act: -0.0001

Data Index: 320410 (C4 (Web Text))

Max Activating Token Index: 936

Text #9

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1591. Min Act: -0.0001

Data Index: 290089 (C4 (Web Text))

Max Activating Token Index: 586

Text #10

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1585. Min Act: -0.0001

Data Index: 43952 (C4 (Web Text))

Max Activating Token Index: 976

Text #11

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1572. Min Act: -0.0001

Data Index: 378869 (C4 (Web Text))

Max Activating Token Index: 131

Text #12

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1460. Min Act: -0.0001

Data Index: 894963 (C4 (Web Text))

Max Activating Token Index: 90

Text #13

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1526. Min Act: -0.0001

Data Index: 860371 (C4 (Web Text))

Max Activating Token Index: 968

Text #14

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1411. Min Act: -0.0001

Data Index: 1163225 (C4 (Web Text))

Max Activating Token Index: 1015

Text #15

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1411. Min Act: -0.0001

Data Index: 512973 (C4 (Web Text))

Max Activating Token Index: 964

Text #16

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1554. Min Act: -0.0001

Data Index: 625697 (C4 (Web Text))

Max Activating Token Index: 239

Text #17

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1360. Min Act: -0.0001

Data Index: 493402 (C4 (Web Text))

Max Activating Token Index: 333

Text #18

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1443. Min Act: -0.0001

Data Index: 926523 (C4 (Web Text))

Max Activating Token Index: 259

Text #19

Max Range: 0.1670. Min Range: -0.1670

Max Act: 0.1545. Min Act: -0.0001

Data Index: 1057908 (C4 (Web Text))

Max Activating Token Index: 270