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

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

Neuron 2688 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0752. Min Act: -0.0001

Data Index: 1201679 (C4 (Web Text))

Max Activating Token Index: 719

Text #1

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0714. Min Act: -0.0001

Data Index: 216459 (C4 (Web Text))

Max Activating Token Index: 414

Text #2

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0591. Min Act: -0.0001

Data Index: 713837 (C4 (Web Text))

Max Activating Token Index: 796

Text #3

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0656. Min Act: -0.0001

Data Index: 30389 (C4 (Web Text))

Max Activating Token Index: 1015

Text #4

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0527. Min Act: -0.0001

Data Index: 550711 (C4 (Web Text))

Max Activating Token Index: 154

Text #5

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0546. Min Act: -0.0001

Data Index: 970797 (C4 (Web Text))

Max Activating Token Index: 346

Text #6

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0529. Min Act: -0.0001

Data Index: 1011677 (C4 (Web Text))

Max Activating Token Index: 547

Text #7

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0506. Min Act: -0.0001

Data Index: 558143 (C4 (Web Text))

Max Activating Token Index: 472

Text #8

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0547. Min Act: -0.0001

Data Index: 643215 (C4 (Web Text))

Max Activating Token Index: 409

Text #9

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0481. Min Act: -0.0000

Data Index: 600710 (C4 (Web Text))

Max Activating Token Index: 20

Text #10

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0502. Min Act: -0.0000

Data Index: 1500958 (Python Code)

Max Activating Token Index: 731

Text #11

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0477. Min Act: -0.0001

Data Index: 106535 (C4 (Web Text))

Max Activating Token Index: 204

Text #12

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0491. Min Act: -0.0001

Data Index: 153142 (C4 (Web Text))

Max Activating Token Index: 639

Text #13

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0434. Min Act: -0.0001

Data Index: 986508 (C4 (Web Text))

Max Activating Token Index: 909

Text #14

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0447. Min Act: -0.0000

Data Index: 670376 (C4 (Web Text))

Max Activating Token Index: 36

Text #15

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0406. Min Act: -0.0000

Data Index: 649967 (C4 (Web Text))

Max Activating Token Index: 772

Text #16

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0406. Min Act: -0.0001

Data Index: 1555777 (Python Code)

Max Activating Token Index: 837

Text #17

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0433. Min Act: -0.0001

Data Index: 875849 (C4 (Web Text))

Max Activating Token Index: 570

Text #18

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0409. Min Act: -0.0001

Data Index: 4688 (C4 (Web Text))

Max Activating Token Index: 295

Text #19

Max Range: 0.0752. Min Range: -0.0752

Max Act: 0.0411. Min Act: -0.0000

Data Index: 1449647 (Python Code)

Max Activating Token Index: 509