Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 817 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0518. Min Act: -0.0001

Data Index: 620459 (C4 (Web Text))

Max Activating Token Index: 586

Text #1

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0501. Min Act: -0.0001

Data Index: 133910 (C4 (Web Text))

Max Activating Token Index: 336

Text #2

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0501. Min Act: -0.0001

Data Index: 1232099 (C4 (Web Text))

Max Activating Token Index: 676

Text #3

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0485. Min Act: -0.0002

Data Index: 1378501 (Python Code)

Max Activating Token Index: 304

Text #4

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0483. Min Act: -0.0002

Data Index: 9788 (C4 (Web Text))

Max Activating Token Index: 504

Text #5

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0456. Min Act: -0.0002

Data Index: 507313 (C4 (Web Text))

Max Activating Token Index: 919

Text #6

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0470. Min Act: -0.0001

Data Index: 427443 (C4 (Web Text))

Max Activating Token Index: 36

Text #7

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0456. Min Act: -0.0001

Data Index: 821187 (C4 (Web Text))

Max Activating Token Index: 159

Text #8

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0455. Min Act: -0.0001

Data Index: 222694 (C4 (Web Text))

Max Activating Token Index: 231

Text #9

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0457. Min Act: -0.0002

Data Index: 888703 (C4 (Web Text))

Max Activating Token Index: 128

Text #10

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0444. Min Act: -0.0002

Data Index: 969090 (C4 (Web Text))

Max Activating Token Index: 978

Text #11

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0426. Min Act: -0.0002

Data Index: 690262 (C4 (Web Text))

Max Activating Token Index: 929

Text #12

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0443. Min Act: -0.0002

Data Index: 104937 (C4 (Web Text))

Max Activating Token Index: 138

Text #13

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0438. Min Act: -0.0002

Data Index: 648758 (C4 (Web Text))

Max Activating Token Index: 1003

Text #14

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0437. Min Act: -0.0002

Data Index: 1156362 (C4 (Web Text))

Max Activating Token Index: 130

Text #15

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0441. Min Act: -0.0002

Data Index: 1576525 (Python Code)

Max Activating Token Index: 100

Text #16

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0427. Min Act: -0.0002

Data Index: 461930 (C4 (Web Text))

Max Activating Token Index: 1022

Text #17

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0427. Min Act: -0.0002

Data Index: 1532575 (Python Code)

Max Activating Token Index: 932

Text #18

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0425. Min Act: -0.0002

Data Index: 983682 (C4 (Web Text))

Max Activating Token Index: 481

Text #19

Max Range: 0.0518. Min Range: -0.0518

Max Act: 0.0443. Min Act: -0.0002

Data Index: 1516065 (Python Code)

Max Activating Token Index: 806