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

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

Neuron 2749 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: 0.1992. Min Range: -0.1992

Max Act: 0.1992. Min Act: -0.0001

Data Index: 212578 (C4 (Web Text))

Max Activating Token Index: 432

Text #1

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1805. Min Act: -0.0001

Data Index: 227151 (C4 (Web Text))

Max Activating Token Index: 347

Text #2

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1713. Min Act: -0.0001

Data Index: 1170987 (C4 (Web Text))

Max Activating Token Index: 129

Text #3

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1645. Min Act: -0.0001

Data Index: 1075972 (C4 (Web Text))

Max Activating Token Index: 607

Text #4

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1682. Min Act: -0.0001

Data Index: 1181478 (C4 (Web Text))

Max Activating Token Index: 309

Text #5

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1700. Min Act: -0.0001

Data Index: 1282558 (C4 (Web Text))

Max Activating Token Index: 742

Text #6

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1530. Min Act: -0.0001

Data Index: 826591 (C4 (Web Text))

Max Activating Token Index: 946

Text #7

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1512. Min Act: -0.0001

Data Index: 459339 (C4 (Web Text))

Max Activating Token Index: 89

Text #8

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1515. Min Act: -0.0001

Data Index: 708452 (C4 (Web Text))

Max Activating Token Index: 484

Text #9

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1561. Min Act: -0.0001

Data Index: 1235167 (C4 (Web Text))

Max Activating Token Index: 597

Text #10

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1435. Min Act: -0.0001

Data Index: 214892 (C4 (Web Text))

Max Activating Token Index: 147

Text #11

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1481. Min Act: -0.0001

Data Index: 116280 (C4 (Web Text))

Max Activating Token Index: 293

Text #12

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1405. Min Act: -0.0001

Data Index: 470488 (C4 (Web Text))

Max Activating Token Index: 961

Text #13

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1384. Min Act: -0.0001

Data Index: 1619173 (Python Code)

Max Activating Token Index: 371

Text #14

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1388. Min Act: -0.0001

Data Index: 427579 (C4 (Web Text))

Max Activating Token Index: 406

Text #15

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1380. Min Act: -0.0001

Data Index: 1416584 (Python Code)

Max Activating Token Index: 437

Text #16

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1424. Min Act: -0.0001

Data Index: 970846 (C4 (Web Text))

Max Activating Token Index: 780

Text #17

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1312. Min Act: -0.0001

Data Index: 673189 (C4 (Web Text))

Max Activating Token Index: 279

Text #18

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1399. Min Act: -0.0001

Data Index: 1697 (C4 (Web Text))

Max Activating Token Index: 406

Text #19

Max Range: 0.1992. Min Range: -0.1992

Max Act: 0.1395. Min Act: -0.0001

Data Index: 1084958 (C4 (Web Text))

Max Activating Token Index: 706