Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 591 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0876. Min Act: -0.0003

Data Index: 1341036 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0606. Min Act: -0.0003

Data Index: 262462 (C4 (Web Text))

Max Activating Token Index: 941

Text #2

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0596. Min Act: -0.0003

Data Index: 229061 (C4 (Web Text))

Max Activating Token Index: 758

Text #3

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0589. Min Act: -0.0003

Data Index: 961940 (C4 (Web Text))

Max Activating Token Index: 163

Text #4

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0576. Min Act: -0.0003

Data Index: 1348442 (C4 (Web Text))

Max Activating Token Index: 748

Text #5

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0557. Min Act: -0.0003

Data Index: 193218 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0546. Min Act: -0.0004

Data Index: 1163503 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0540. Min Act: -0.0003

Data Index: 1237464 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0524. Min Act: -0.0003

Data Index: 724813 (C4 (Web Text))

Max Activating Token Index: 220

Text #9

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0506. Min Act: -0.0003

Data Index: 1633928 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0505. Min Act: -0.0003

Data Index: 595763 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0504. Min Act: -0.0003

Data Index: 737538 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0503. Min Act: -0.0004

Data Index: 42042 (C4 (Web Text))

Max Activating Token Index: 57

Text #13

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0486. Min Act: -0.0003

Data Index: 976045 (C4 (Web Text))

Max Activating Token Index: 704

Text #14

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0481. Min Act: -0.0003

Data Index: 78796 (C4 (Web Text))

Max Activating Token Index: 319

Text #15

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0471. Min Act: -0.0003

Data Index: 506970 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0474. Min Act: -0.0003

Data Index: 1201039 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0451. Min Act: -0.0003

Data Index: 857329 (C4 (Web Text))

Max Activating Token Index: 716

Text #18

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0468. Min Act: -0.0003

Data Index: 725172 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.0876. Min Range: -0.0876

Max Act: 0.0457. Min Act: -0.0003

Data Index: 1637674 (Python Code)

Max Activating Token Index: 1023