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

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

Neuron 3236 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1927. Min Range: -2.1927

Max Act: 2.1927. Min Act: -0.0000

Data Index: 1594266 (Python Code)

Max Activating Token Index: 300

Text #1

Max Range: 2.1927. Min Range: -2.1927

Max Act: 2.0126. Min Act: -0.0000

Data Index: 860747 (C4 (Web Text))

Max Activating Token Index: 170

Text #2

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.8955. Min Act: -0.0000

Data Index: 139330 (C4 (Web Text))

Max Activating Token Index: 89

Text #3

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7728. Min Act: -0.0000

Data Index: 1447232 (Python Code)

Max Activating Token Index: 758

Text #4

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.8158. Min Act: -0.0000

Data Index: 17837 (C4 (Web Text))

Max Activating Token Index: 607

Text #5

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.8136. Min Act: -0.0000

Data Index: 1279153 (C4 (Web Text))

Max Activating Token Index: 208

Text #6

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7542. Min Act: -0.0000

Data Index: 1089836 (C4 (Web Text))

Max Activating Token Index: 817

Text #7

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7534. Min Act: -0.0000

Data Index: 42981 (C4 (Web Text))

Max Activating Token Index: 309

Text #8

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7329. Min Act: -0.0000

Data Index: 280607 (C4 (Web Text))

Max Activating Token Index: 67

Text #9

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.6489. Min Act: -0.0000

Data Index: 1429808 (Python Code)

Max Activating Token Index: 670

Text #10

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7418. Min Act: -0.0000

Data Index: 1073218 (C4 (Web Text))

Max Activating Token Index: 758

Text #11

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7085. Min Act: -0.0000

Data Index: 520324 (C4 (Web Text))

Max Activating Token Index: 408

Text #12

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7020. Min Act: -0.0000

Data Index: 143201 (C4 (Web Text))

Max Activating Token Index: 39

Text #13

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.6960. Min Act: -0.0000

Data Index: 512699 (C4 (Web Text))

Max Activating Token Index: 77

Text #14

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.7813. Min Act: -0.0000

Data Index: 1433302 (Python Code)

Max Activating Token Index: 989

Text #15

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.5876. Min Act: -0.0000

Data Index: 1488843 (Python Code)

Max Activating Token Index: 248

Text #16

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.6753. Min Act: -0.0000

Data Index: 457310 (C4 (Web Text))

Max Activating Token Index: 981

Text #17

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.6704. Min Act: -0.0000

Data Index: 801872 (C4 (Web Text))

Max Activating Token Index: 124

Text #18

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.5810. Min Act: -0.0000

Data Index: 1622973 (Python Code)

Max Activating Token Index: 768

Text #19

Max Range: 2.1927. Min Range: -2.1927

Max Act: 1.5767. Min Act: -0.0000

Data Index: 1165700 (C4 (Web Text))

Max Activating Token Index: 846