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

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

Neuron 1225 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2959. Min Act: -0.0001

Data Index: 718346 (C4 (Web Text))

Max Activating Token Index: 511

Text #1

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.3039. Min Act: -0.0001

Data Index: 1332077 (C4 (Web Text))

Max Activating Token Index: 465

Text #2

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2834. Min Act: -0.0001

Data Index: 160652 (C4 (Web Text))

Max Activating Token Index: 154

Text #3

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2702. Min Act: -0.0001

Data Index: 1160130 (C4 (Web Text))

Max Activating Token Index: 975

Text #4

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2904. Min Act: -0.0001

Data Index: 437544 (C4 (Web Text))

Max Activating Token Index: 889

Text #5

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2551. Min Act: -0.0001

Data Index: 1170628 (C4 (Web Text))

Max Activating Token Index: 804

Text #6

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2709. Min Act: -0.0001

Data Index: 1176039 (C4 (Web Text))

Max Activating Token Index: 254

Text #7

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2731. Min Act: -0.0000

Data Index: 692899 (C4 (Web Text))

Max Activating Token Index: 678

Text #8

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2612. Min Act: -0.0000

Data Index: 29887 (C4 (Web Text))

Max Activating Token Index: 116

Text #9

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2485. Min Act: -0.0000

Data Index: 665093 (C4 (Web Text))

Max Activating Token Index: 9

Text #10

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2633. Min Act: -0.0001

Data Index: 194316 (C4 (Web Text))

Max Activating Token Index: 760

Text #11

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2508. Min Act: -0.0001

Data Index: 223925 (C4 (Web Text))

Max Activating Token Index: 307

Text #12

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2556. Min Act: -0.0000

Data Index: 1492132 (Python Code)

Max Activating Token Index: 6

Text #13

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2470. Min Act: -0.0000

Data Index: 1024587 (C4 (Web Text))

Max Activating Token Index: 182

Text #14

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2530. Min Act: -0.0000

Data Index: 216506 (C4 (Web Text))

Max Activating Token Index: 647

Text #15

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2450. Min Act: -0.0000

Data Index: 566139 (C4 (Web Text))

Max Activating Token Index: 903

Text #16

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2425. Min Act: -0.0001

Data Index: 195273 (C4 (Web Text))

Max Activating Token Index: 23

Text #17

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2502. Min Act: -0.0001

Data Index: 25779 (C4 (Web Text))

Max Activating Token Index: 919

Text #18

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2419. Min Act: -0.0001

Data Index: 1286624 (C4 (Web Text))

Max Activating Token Index: 601

Text #19

Max Range: 0.3039. Min Range: -0.3039

Max Act: 0.2388. Min Act: -0.0000

Data Index: 1216892 (C4 (Web Text))

Max Activating Token Index: 827