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

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

Neuron 3586 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0661. Min Act: -0.0002

Data Index: 1241226 (C4 (Web Text))

Max Activating Token Index: 32

Text #1

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0660. Min Act: -0.0002

Data Index: 900365 (C4 (Web Text))

Max Activating Token Index: 522

Text #2

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0646. Min Act: -0.0002

Data Index: 434491 (C4 (Web Text))

Max Activating Token Index: 563

Text #3

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0639. Min Act: -0.0002

Data Index: 1610874 (Python Code)

Max Activating Token Index: 703

Text #4

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0641. Min Act: -0.0002

Data Index: 957222 (C4 (Web Text))

Max Activating Token Index: 19

Text #5

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0594. Min Act: -0.0002

Data Index: 1411427 (Python Code)

Max Activating Token Index: 890

Text #6

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0612. Min Act: -0.0002

Data Index: 1165725 (C4 (Web Text))

Max Activating Token Index: 209

Text #7

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0601. Min Act: -0.0002

Data Index: 674043 (C4 (Web Text))

Max Activating Token Index: 818

Text #8

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0605. Min Act: -0.0002

Data Index: 1469600 (Python Code)

Max Activating Token Index: 899

Text #9

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0588. Min Act: -0.0002

Data Index: 1481582 (Python Code)

Max Activating Token Index: 576

Text #10

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0597. Min Act: -0.0002

Data Index: 1638406 (Python Code)

Max Activating Token Index: 440

Text #11

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0579. Min Act: -0.0002

Data Index: 491982 (C4 (Web Text))

Max Activating Token Index: 94

Text #12

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0568. Min Act: -0.0002

Data Index: 1542203 (Python Code)

Max Activating Token Index: 885

Text #13

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0583. Min Act: -0.0002

Data Index: 240067 (C4 (Web Text))

Max Activating Token Index: 28

Text #14

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0576. Min Act: -0.0002

Data Index: 1619459 (Python Code)

Max Activating Token Index: 600

Text #15

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0573. Min Act: -0.0002

Data Index: 212739 (C4 (Web Text))

Max Activating Token Index: 254

Text #16

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0562. Min Act: -0.0002

Data Index: 952332 (C4 (Web Text))

Max Activating Token Index: 72

Text #17

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0575. Min Act: -0.0003

Data Index: 1057625 (C4 (Web Text))

Max Activating Token Index: 800

Text #18

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0554. Min Act: -0.0002

Data Index: 585139 (C4 (Web Text))

Max Activating Token Index: 193

Text #19

Max Range: 0.0661. Min Range: -0.0661

Max Act: 0.0547. Min Act: -0.0002

Data Index: 351203 (C4 (Web Text))

Max Activating Token Index: 510