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

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

Neuron 4848 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0657. Min Range: -3.0657

Max Act: 3.0657. Min Act: -0.0000

Data Index: 1419705 (Python Code)

Max Activating Token Index: 786

Text #1

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.7512. Min Act: -0.0000

Data Index: 1253972 (C4 (Web Text))

Max Activating Token Index: 204

Text #2

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.7407. Min Act: -0.0000

Data Index: 1167155 (C4 (Web Text))

Max Activating Token Index: 331

Text #3

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.7620. Min Act: -0.0000

Data Index: 1049107 (C4 (Web Text))

Max Activating Token Index: 198

Text #4

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.5106. Min Act: -0.0000

Data Index: 611638 (C4 (Web Text))

Max Activating Token Index: 291

Text #5

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.6249. Min Act: -0.0000

Data Index: 453665 (C4 (Web Text))

Max Activating Token Index: 522

Text #6

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.5805. Min Act: -0.0000

Data Index: 1471114 (Python Code)

Max Activating Token Index: 752

Text #7

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.5181. Min Act: -0.0000

Data Index: 1505246 (Python Code)

Max Activating Token Index: 948

Text #8

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.5616. Min Act: -0.0000

Data Index: 1394801 (Python Code)

Max Activating Token Index: 523

Text #9

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.3594. Min Act: -0.0000

Data Index: 1500323 (Python Code)

Max Activating Token Index: 848

Text #10

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.2626. Min Act: -0.0000

Data Index: 1531230 (Python Code)

Max Activating Token Index: 108

Text #11

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.1355. Min Act: -0.0000

Data Index: 1448660 (Python Code)

Max Activating Token Index: 675

Text #12

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.0416. Min Act: -0.0000

Data Index: 269954 (C4 (Web Text))

Max Activating Token Index: 399

Text #13

Max Range: 3.0657. Min Range: -3.0657

Max Act: 1.9525. Min Act: -0.0000

Data Index: 1578765 (Python Code)

Max Activating Token Index: 397

Text #14

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.0559. Min Act: -0.0000

Data Index: 397854 (C4 (Web Text))

Max Activating Token Index: 421

Text #15

Max Range: 3.0657. Min Range: -3.0657

Max Act: 1.8286. Min Act: -0.0000

Data Index: 676911 (C4 (Web Text))

Max Activating Token Index: 119

Text #16

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.0327. Min Act: -0.0000

Data Index: 1540006 (Python Code)

Max Activating Token Index: 496

Text #17

Max Range: 3.0657. Min Range: -3.0657

Max Act: 1.8831. Min Act: -0.0000

Data Index: 1213543 (C4 (Web Text))

Max Activating Token Index: 618

Text #18

Max Range: 3.0657. Min Range: -3.0657

Max Act: 2.0098. Min Act: -0.0000

Data Index: 1650949 (Python Code)

Max Activating Token Index: 617

Text #19

Max Range: 3.0657. Min Range: -3.0657

Max Act: 1.9777. Min Act: -0.0000

Data Index: 1593517 (Python Code)

Max Activating Token Index: 106