Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3524 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1983. Min Act: -0.0001

Data Index: 727662 (C4 (Web Text))

Max Activating Token Index: 669

Text #1

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1819. Min Act: -0.0001

Data Index: 200798 (C4 (Web Text))

Max Activating Token Index: 380

Text #2

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1545. Min Act: -0.0001

Data Index: 346944 (C4 (Web Text))

Max Activating Token Index: 423

Text #3

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1799. Min Act: -0.0001

Data Index: 317090 (C4 (Web Text))

Max Activating Token Index: 396

Text #4

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1647. Min Act: -0.0001

Data Index: 1619076 (Python Code)

Max Activating Token Index: 150

Text #5

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1564. Min Act: -0.0001

Data Index: 1530355 (Python Code)

Max Activating Token Index: 208

Text #6

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1503. Min Act: -0.0001

Data Index: 108603 (C4 (Web Text))

Max Activating Token Index: 606

Text #7

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1484. Min Act: -0.0001

Data Index: 1454420 (Python Code)

Max Activating Token Index: 154

Text #8

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1515. Min Act: -0.0001

Data Index: 976751 (C4 (Web Text))

Max Activating Token Index: 74

Text #9

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1558. Min Act: -0.0001

Data Index: 1491234 (Python Code)

Max Activating Token Index: 268

Text #10

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1364. Min Act: -0.0001

Data Index: 1561387 (Python Code)

Max Activating Token Index: 234

Text #11

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1547. Min Act: -0.0001

Data Index: 543488 (C4 (Web Text))

Max Activating Token Index: 506

Text #12

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1500. Min Act: -0.0001

Data Index: 1642938 (Python Code)

Max Activating Token Index: 618

Text #13

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1589. Min Act: -0.0001

Data Index: 1388547 (Python Code)

Max Activating Token Index: 92

Text #14

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1581. Min Act: -0.0001

Data Index: 1453315 (Python Code)

Max Activating Token Index: 336

Text #15

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1487. Min Act: -0.0001

Data Index: 2867 (C4 (Web Text))

Max Activating Token Index: 617

Text #16

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1374. Min Act: -0.0001

Data Index: 1594132 (Python Code)

Max Activating Token Index: 450

Text #17

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1522. Min Act: -0.0001

Data Index: 1476032 (Python Code)

Max Activating Token Index: 108

Text #18

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1543. Min Act: -0.0001

Data Index: 1397481 (Python Code)

Max Activating Token Index: 446

Text #19

Max Range: 0.1983. Min Range: -0.1983

Max Act: 0.1449. Min Act: -0.0001

Data Index: 266776 (C4 (Web Text))

Max Activating Token Index: 712