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

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

Neuron 5752 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.8049. Min Act: -0.0001

Data Index: 762674 (C4 (Web Text))

Max Activating Token Index: 867

Text #1

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6103. Min Act: -0.0001

Data Index: 1023193 (C4 (Web Text))

Max Activating Token Index: 401

Text #2

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6162. Min Act: -0.0001

Data Index: 93283 (C4 (Web Text))

Max Activating Token Index: 730

Text #3

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6499. Min Act: -0.0001

Data Index: 801254 (C4 (Web Text))

Max Activating Token Index: 913

Text #4

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6611. Min Act: -0.0001

Data Index: 1570945 (Python Code)

Max Activating Token Index: 889

Text #5

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6240. Min Act: -0.0001

Data Index: 663887 (C4 (Web Text))

Max Activating Token Index: 262

Text #6

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5748. Min Act: -0.0001

Data Index: 1453959 (Python Code)

Max Activating Token Index: 61

Text #7

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6135. Min Act: -0.0001

Data Index: 415523 (C4 (Web Text))

Max Activating Token Index: 720

Text #8

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5949. Min Act: -0.0001

Data Index: 698617 (C4 (Web Text))

Max Activating Token Index: 680

Text #9

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5770. Min Act: -0.0001

Data Index: 269836 (C4 (Web Text))

Max Activating Token Index: 995

Text #10

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6639. Min Act: -0.0001

Data Index: 18455 (C4 (Web Text))

Max Activating Token Index: 333

Text #11

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5721. Min Act: -0.0001

Data Index: 1516119 (Python Code)

Max Activating Token Index: 352

Text #12

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5693. Min Act: -0.0001

Data Index: 744768 (C4 (Web Text))

Max Activating Token Index: 500

Text #13

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5554. Min Act: -0.0001

Data Index: 1040950 (C4 (Web Text))

Max Activating Token Index: 84

Text #14

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.6533. Min Act: -0.0001

Data Index: 1331366 (C4 (Web Text))

Max Activating Token Index: 853

Text #15

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5753. Min Act: -0.0001

Data Index: 220835 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5424. Min Act: -0.0001

Data Index: 224071 (C4 (Web Text))

Max Activating Token Index: 805

Text #17

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5735. Min Act: -0.0001

Data Index: 1063285 (C4 (Web Text))

Max Activating Token Index: 62

Text #18

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5669. Min Act: -0.0001

Data Index: 472354 (C4 (Web Text))

Max Activating Token Index: 630

Text #19

Max Range: 0.8049. Min Range: -0.8049

Max Act: 0.5595. Min Act: -0.0001

Data Index: 408791 (C4 (Web Text))

Max Activating Token Index: 769