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

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

Neuron 3301 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.6194. Min Act: -0.0000

Data Index: 1607177 (Python Code)

Max Activating Token Index: 962

Text #1

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.6817. Min Act: -0.0000

Data Index: 1367384 (Python Code)

Max Activating Token Index: 73

Text #2

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.6187. Min Act: -0.0001

Data Index: 1467569 (Python Code)

Max Activating Token Index: 253

Text #3

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.6048. Min Act: -0.0000

Data Index: 1598116 (Python Code)

Max Activating Token Index: 42

Text #4

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.5881. Min Act: -0.0000

Data Index: 1095557 (C4 (Web Text))

Max Activating Token Index: 190

Text #5

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4525. Min Act: -0.0000

Data Index: 1449529 (Python Code)

Max Activating Token Index: 830

Text #6

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.5385. Min Act: -0.0000

Data Index: 1619158 (Python Code)

Max Activating Token Index: 542

Text #7

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4131. Min Act: -0.0000

Data Index: 59213 (C4 (Web Text))

Max Activating Token Index: 496

Text #8

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4846. Min Act: -0.0000

Data Index: 1368493 (Python Code)

Max Activating Token Index: 911

Text #9

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4127. Min Act: -0.0000

Data Index: 1041724 (C4 (Web Text))

Max Activating Token Index: 16

Text #10

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4099. Min Act: -0.0000

Data Index: 65203 (C4 (Web Text))

Max Activating Token Index: 289

Text #11

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.2726. Min Act: -0.0000

Data Index: 1272392 (C4 (Web Text))

Max Activating Token Index: 336

Text #12

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4007. Min Act: -0.0001

Data Index: 1609193 (Python Code)

Max Activating Token Index: 277

Text #13

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.3700. Min Act: -0.0000

Data Index: 1563735 (Python Code)

Max Activating Token Index: 119

Text #14

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.2487. Min Act: -0.0000

Data Index: 1550069 (Python Code)

Max Activating Token Index: 92

Text #15

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4550. Min Act: -0.0000

Data Index: 906108 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.1951. Min Act: -0.0000

Data Index: 1370344 (Python Code)

Max Activating Token Index: 765

Text #17

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.4332. Min Act: -0.0000

Data Index: 1603837 (Python Code)

Max Activating Token Index: 843

Text #18

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.2819. Min Act: -0.0000

Data Index: 519671 (C4 (Web Text))

Max Activating Token Index: 502

Text #19

Max Range: 2.6817. Min Range: -2.6817

Max Act: 2.2567. Min Act: -0.0000

Data Index: 1513377 (Python Code)

Max Activating Token Index: 926