Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1534 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4306. Min Act: -0.0000

Data Index: 829475 (C4 (Web Text))

Max Activating Token Index: 225

Text #1

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4287. Min Act: -0.0000

Data Index: 1145762 (C4 (Web Text))

Max Activating Token Index: 928

Text #2

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4227. Min Act: -0.0000

Data Index: 696794 (C4 (Web Text))

Max Activating Token Index: 18

Text #3

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4072. Min Act: -0.0000

Data Index: 239117 (C4 (Web Text))

Max Activating Token Index: 57

Text #4

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4047. Min Act: -0.0000

Data Index: 1589812 (Python Code)

Max Activating Token Index: 543

Text #5

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4348. Min Act: -0.0000

Data Index: 1474116 (Python Code)

Max Activating Token Index: 152

Text #6

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3991. Min Act: -0.0000

Data Index: 1141631 (C4 (Web Text))

Max Activating Token Index: 882

Text #7

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4004. Min Act: -0.0000

Data Index: 1446759 (Python Code)

Max Activating Token Index: 484

Text #8

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3897. Min Act: -0.0000

Data Index: 1648707 (Python Code)

Max Activating Token Index: 966

Text #9

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.4132. Min Act: -0.0000

Data Index: 632585 (C4 (Web Text))

Max Activating Token Index: 652

Text #10

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3834. Min Act: -0.0000

Data Index: 1428734 (Python Code)

Max Activating Token Index: 66

Text #11

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3974. Min Act: -0.0000

Data Index: 1183107 (C4 (Web Text))

Max Activating Token Index: 305

Text #12

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3800. Min Act: -0.0000

Data Index: 1145941 (C4 (Web Text))

Max Activating Token Index: 536

Text #13

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3682. Min Act: -0.0000

Data Index: 237099 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3506. Min Act: -0.0000

Data Index: 513965 (C4 (Web Text))

Max Activating Token Index: 47

Text #15

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3462. Min Act: -0.0000

Data Index: 276890 (C4 (Web Text))

Max Activating Token Index: 208

Text #16

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3403. Min Act: -0.0000

Data Index: 419319 (C4 (Web Text))

Max Activating Token Index: 480

Text #17

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3417. Min Act: -0.0000

Data Index: 553292 (C4 (Web Text))

Max Activating Token Index: 349

Text #18

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3377. Min Act: -0.0000

Data Index: 1520716 (Python Code)

Max Activating Token Index: 622

Text #19

Max Range: 0.4348. Min Range: -0.4348

Max Act: 0.3145. Min Act: -0.0000

Data Index: 1607322 (Python Code)

Max Activating Token Index: 114