Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5496 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6801. Min Act: -0.0000

Data Index: 204375 (C4 (Web Text))

Max Activating Token Index: 700

Text #1

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6577. Min Act: -0.0000

Data Index: 1103982 (C4 (Web Text))

Max Activating Token Index: 738

Text #2

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6562. Min Act: -0.0001

Data Index: 1298266 (C4 (Web Text))

Max Activating Token Index: 262

Text #3

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6340. Min Act: -0.0000

Data Index: 387316 (C4 (Web Text))

Max Activating Token Index: 529

Text #4

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6104. Min Act: -0.0000

Data Index: 1454538 (Python Code)

Max Activating Token Index: 241

Text #5

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6433. Min Act: -0.0001

Data Index: 995127 (C4 (Web Text))

Max Activating Token Index: 736

Text #6

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5822. Min Act: -0.0000

Data Index: 1567553 (Python Code)

Max Activating Token Index: 992

Text #7

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6042. Min Act: -0.0001

Data Index: 953585 (C4 (Web Text))

Max Activating Token Index: 423

Text #8

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5748. Min Act: -0.0000

Data Index: 431363 (C4 (Web Text))

Max Activating Token Index: 163

Text #9

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6233. Min Act: -0.0000

Data Index: 1494746 (Python Code)

Max Activating Token Index: 271

Text #10

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5896. Min Act: -0.0000

Data Index: 1593373 (Python Code)

Max Activating Token Index: 999

Text #11

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6312. Min Act: -0.0000

Data Index: 74132 (C4 (Web Text))

Max Activating Token Index: 905

Text #12

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.4993. Min Act: -0.0000

Data Index: 875286 (C4 (Web Text))

Max Activating Token Index: 944

Text #13

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5687. Min Act: -0.0000

Data Index: 1071347 (C4 (Web Text))

Max Activating Token Index: 13

Text #14

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5341. Min Act: -0.0000

Data Index: 332168 (C4 (Web Text))

Max Activating Token Index: 807

Text #15

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.4988. Min Act: -0.0000

Data Index: 1430746 (Python Code)

Max Activating Token Index: 214

Text #16

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5452. Min Act: -0.0000

Data Index: 1045566 (C4 (Web Text))

Max Activating Token Index: 581

Text #17

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5431. Min Act: -0.0000

Data Index: 1627309 (Python Code)

Max Activating Token Index: 194

Text #18

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.6013. Min Act: -0.0000

Data Index: 1060271 (C4 (Web Text))

Max Activating Token Index: 370

Text #19

Max Range: 0.6801. Min Range: -0.6801

Max Act: 0.5579. Min Act: -0.0000

Data Index: 855737 (C4 (Web Text))

Max Activating Token Index: 10