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

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

Neuron 551 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.4751. Min Act: -0.0001

Data Index: 1340573 (C4 (Web Text))

Max Activating Token Index: 351

Text #1

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.4962. Min Act: -0.0001

Data Index: 1057416 (C4 (Web Text))

Max Activating Token Index: 974

Text #2

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.4472. Min Act: -0.0001

Data Index: 1444643 (Python Code)

Max Activating Token Index: 777

Text #3

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.4243. Min Act: -0.0000

Data Index: 1381490 (Python Code)

Max Activating Token Index: 284

Text #4

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.4127. Min Act: -0.0000

Data Index: 1387131 (Python Code)

Max Activating Token Index: 239

Text #5

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3995. Min Act: -0.0001

Data Index: 28158 (C4 (Web Text))

Max Activating Token Index: 1018

Text #6

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3736. Min Act: -0.0001

Data Index: 54442 (C4 (Web Text))

Max Activating Token Index: 567

Text #7

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3688. Min Act: -0.0000

Data Index: 1573398 (Python Code)

Max Activating Token Index: 831

Text #8

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3659. Min Act: -0.0001

Data Index: 1238845 (C4 (Web Text))

Max Activating Token Index: 536

Text #9

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3873. Min Act: -0.0000

Data Index: 1376659 (Python Code)

Max Activating Token Index: 986

Text #10

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3618. Min Act: -0.0001

Data Index: 986249 (C4 (Web Text))

Max Activating Token Index: 260

Text #11

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3700. Min Act: -0.0001

Data Index: 905421 (C4 (Web Text))

Max Activating Token Index: 248

Text #12

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3417. Min Act: -0.0001

Data Index: 631802 (C4 (Web Text))

Max Activating Token Index: 981

Text #13

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3625. Min Act: -0.0001

Data Index: 387862 (C4 (Web Text))

Max Activating Token Index: 127

Text #14

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3640. Min Act: -0.0001

Data Index: 213018 (C4 (Web Text))

Max Activating Token Index: 332

Text #15

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3244. Min Act: -0.0000

Data Index: 1497638 (Python Code)

Max Activating Token Index: 952

Text #16

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3428. Min Act: -0.0001

Data Index: 724542 (C4 (Web Text))

Max Activating Token Index: 453

Text #17

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3337. Min Act: -0.0001

Data Index: 1392590 (Python Code)

Max Activating Token Index: 244

Text #18

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3695. Min Act: -0.0001

Data Index: 1575052 (Python Code)

Max Activating Token Index: 784

Text #19

Max Range: 0.4962. Min Range: -0.4962

Max Act: 0.3478. Min Act: -0.0000

Data Index: 1597086 (Python Code)

Max Activating Token Index: 969