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

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

Neuron 4534 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0521. Min Act: -0.0002

Data Index: 1639929 (Python Code)

Max Activating Token Index: 129

Text #1

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0404. Min Act: -0.0002

Data Index: 473521 (C4 (Web Text))

Max Activating Token Index: 993

Text #2

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0440. Min Act: -0.0002

Data Index: 646828 (C4 (Web Text))

Max Activating Token Index: 104

Text #3

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0448. Min Act: -0.0002

Data Index: 1237496 (C4 (Web Text))

Max Activating Token Index: 400

Text #4

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0411. Min Act: -0.0002

Data Index: 185179 (C4 (Web Text))

Max Activating Token Index: 750

Text #5

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0513. Min Act: -0.0002

Data Index: 1088924 (C4 (Web Text))

Max Activating Token Index: 925

Text #6

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0442. Min Act: -0.0002

Data Index: 1607073 (Python Code)

Max Activating Token Index: 652

Text #7

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0409. Min Act: -0.0002

Data Index: 897967 (C4 (Web Text))

Max Activating Token Index: 114

Text #8

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0409. Min Act: -0.0002

Data Index: 166855 (C4 (Web Text))

Max Activating Token Index: 542

Text #9

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0434. Min Act: -0.0002

Data Index: 357627 (C4 (Web Text))

Max Activating Token Index: 33

Text #10

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0433. Min Act: -0.0002

Data Index: 1633815 (Python Code)

Max Activating Token Index: 25

Text #11

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0434. Min Act: -0.0002

Data Index: 1534890 (Python Code)

Max Activating Token Index: 129

Text #12

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0418. Min Act: -0.0002

Data Index: 1586955 (Python Code)

Max Activating Token Index: 427

Text #13

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0378. Min Act: -0.0002

Data Index: 1076680 (C4 (Web Text))

Max Activating Token Index: 565

Text #14

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0405. Min Act: -0.0002

Data Index: 503739 (C4 (Web Text))

Max Activating Token Index: 652

Text #15

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0377. Min Act: -0.0002

Data Index: 1280733 (C4 (Web Text))

Max Activating Token Index: 185

Text #16

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0403. Min Act: -0.0002

Data Index: 477313 (C4 (Web Text))

Max Activating Token Index: 56

Text #17

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0406. Min Act: -0.0002

Data Index: 1114469 (C4 (Web Text))

Max Activating Token Index: 7

Text #18

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0384. Min Act: -0.0002

Data Index: 28248 (C4 (Web Text))

Max Activating Token Index: 144

Text #19

Max Range: 0.0521. Min Range: -0.0521

Max Act: 0.0366. Min Act: -0.0002

Data Index: 108944 (C4 (Web Text))

Max Activating Token Index: 393