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

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

Neuron 4220 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0543. Min Act: -0.0001

Data Index: 562070 (C4 (Web Text))

Max Activating Token Index: 971

Text #1

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0550. Min Act: -0.0001

Data Index: 1064386 (C4 (Web Text))

Max Activating Token Index: 698

Text #2

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0531. Min Act: -0.0001

Data Index: 1044027 (C4 (Web Text))

Max Activating Token Index: 299

Text #3

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0518. Min Act: -0.0001

Data Index: 124471 (C4 (Web Text))

Max Activating Token Index: 238

Text #4

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0468. Min Act: -0.0001

Data Index: 1036729 (C4 (Web Text))

Max Activating Token Index: 1022

Text #5

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0502. Min Act: -0.0001

Data Index: 417002 (C4 (Web Text))

Max Activating Token Index: 486

Text #6

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0457. Min Act: -0.0001

Data Index: 24759 (C4 (Web Text))

Max Activating Token Index: 437

Text #7

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0474. Min Act: -0.0001

Data Index: 762746 (C4 (Web Text))

Max Activating Token Index: 836

Text #8

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0427. Min Act: -0.0001

Data Index: 1295539 (C4 (Web Text))

Max Activating Token Index: 875

Text #9

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0461. Min Act: -0.0000

Data Index: 1469587 (Python Code)

Max Activating Token Index: 969

Text #10

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0460. Min Act: -0.0000

Data Index: 1617040 (Python Code)

Max Activating Token Index: 961

Text #11

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0448. Min Act: -0.0001

Data Index: 693127 (C4 (Web Text))

Max Activating Token Index: 152

Text #12

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0416. Min Act: -0.0000

Data Index: 1450698 (Python Code)

Max Activating Token Index: 996

Text #13

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0451. Min Act: -0.0001

Data Index: 34428 (C4 (Web Text))

Max Activating Token Index: 890

Text #14

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0449. Min Act: -0.0001

Data Index: 1498620 (Python Code)

Max Activating Token Index: 514

Text #15

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0451. Min Act: -0.0000

Data Index: 1404111 (Python Code)

Max Activating Token Index: 939

Text #16

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0448. Min Act: -0.0000

Data Index: 1434920 (Python Code)

Max Activating Token Index: 993

Text #17

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0434. Min Act: -0.0001

Data Index: 147179 (C4 (Web Text))

Max Activating Token Index: 867

Text #18

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0421. Min Act: -0.0001

Data Index: 685700 (C4 (Web Text))

Max Activating Token Index: 1010

Text #19

Max Range: 0.0550. Min Range: -0.0550

Max Act: 0.0428. Min Act: -0.0001

Data Index: 1425355 (Python Code)

Max Activating Token Index: 272