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

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

Neuron 253 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.2941. Min Act: -0.0001

Data Index: 261076 (C4 (Web Text))

Max Activating Token Index: 15

Text #1

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.2337. Min Act: -0.0001

Data Index: 1289850 (C4 (Web Text))

Max Activating Token Index: 12

Text #2

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.2265. Min Act: -0.0001

Data Index: 705705 (C4 (Web Text))

Max Activating Token Index: 16

Text #3

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.2066. Min Act: -0.0001

Data Index: 1235127 (C4 (Web Text))

Max Activating Token Index: 18

Text #4

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1785. Min Act: -0.0001

Data Index: 276789 (C4 (Web Text))

Max Activating Token Index: 16

Text #5

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1765. Min Act: -0.0001

Data Index: 1069824 (C4 (Web Text))

Max Activating Token Index: 66

Text #6

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1771. Min Act: -0.0001

Data Index: 380228 (C4 (Web Text))

Max Activating Token Index: 235

Text #7

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1709. Min Act: -0.0001

Data Index: 1587944 (Python Code)

Max Activating Token Index: 10

Text #8

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1687. Min Act: -0.0001

Data Index: 1358645 (C4 (Web Text))

Max Activating Token Index: 24

Text #9

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1673. Min Act: -0.0001

Data Index: 162076 (C4 (Web Text))

Max Activating Token Index: 12

Text #10

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1600. Min Act: -0.0001

Data Index: 294818 (C4 (Web Text))

Max Activating Token Index: 19

Text #11

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1605. Min Act: -0.0001

Data Index: 1062058 (C4 (Web Text))

Max Activating Token Index: 42

Text #12

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1577. Min Act: -0.0001

Data Index: 310232 (C4 (Web Text))

Max Activating Token Index: 505

Text #13

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1636. Min Act: -0.0001

Data Index: 1177985 (C4 (Web Text))

Max Activating Token Index: 15

Text #14

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1621. Min Act: -0.0001

Data Index: 1494985 (Python Code)

Max Activating Token Index: 8

Text #15

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1523. Min Act: -0.0001

Data Index: 938681 (C4 (Web Text))

Max Activating Token Index: 230

Text #16

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1417. Min Act: -0.0001

Data Index: 312871 (C4 (Web Text))

Max Activating Token Index: 1017

Text #17

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1573. Min Act: -0.0001

Data Index: 390971 (C4 (Web Text))

Max Activating Token Index: 582

Text #18

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1552. Min Act: -0.0001

Data Index: 318962 (C4 (Web Text))

Max Activating Token Index: 17

Text #19

Max Range: 0.2941. Min Range: -0.2941

Max Act: 0.1521. Min Act: -0.0001

Data Index: 229659 (C4 (Web Text))

Max Activating Token Index: 101