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

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

Neuron 4208 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5591. Min Range: -2.5591

Max Act: 2.3846. Min Act: -0.0000

Data Index: 1650 (C4 (Web Text))

Max Activating Token Index: 196

Text #1

Max Range: 2.5591. Min Range: -2.5591

Max Act: 2.5591. Min Act: -0.0000

Data Index: 38100 (C4 (Web Text))

Max Activating Token Index: 730

Text #2

Max Range: 2.5591. Min Range: -2.5591

Max Act: 2.2610. Min Act: -0.0000

Data Index: 731159 (C4 (Web Text))

Max Activating Token Index: 58

Text #3

Max Range: 2.5591. Min Range: -2.5591

Max Act: 2.2363. Min Act: -0.0000

Data Index: 973138 (C4 (Web Text))

Max Activating Token Index: 400

Text #4

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.9603. Min Act: -0.0000

Data Index: 604592 (C4 (Web Text))

Max Activating Token Index: 290

Text #5

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.9643. Min Act: -0.0000

Data Index: 59651 (C4 (Web Text))

Max Activating Token Index: 900

Text #6

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.9494. Min Act: -0.0000

Data Index: 1066001 (C4 (Web Text))

Max Activating Token Index: 452

Text #7

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.8867. Min Act: -0.0000

Data Index: 1320140 (C4 (Web Text))

Max Activating Token Index: 193

Text #8

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.8503. Min Act: -0.0000

Data Index: 1332603 (C4 (Web Text))

Max Activating Token Index: 741

Text #9

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.4569. Min Act: -0.0000

Data Index: 1581468 (Python Code)

Max Activating Token Index: 657

Text #10

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.7240. Min Act: -0.0000

Data Index: 1211667 (C4 (Web Text))

Max Activating Token Index: 480

Text #11

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.2868. Min Act: -0.0000

Data Index: 1375945 (Python Code)

Max Activating Token Index: 851

Text #12

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.5089. Min Act: -0.0000

Data Index: 912617 (C4 (Web Text))

Max Activating Token Index: 566

Text #13

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.4877. Min Act: -0.0000

Data Index: 742037 (C4 (Web Text))

Max Activating Token Index: 57

Text #14

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.3694. Min Act: -0.0000

Data Index: 1397098 (Python Code)

Max Activating Token Index: 906

Text #15

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.4143. Min Act: -0.0000

Data Index: 1630403 (Python Code)

Max Activating Token Index: 489

Text #16

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.3903. Min Act: -0.0000

Data Index: 390783 (C4 (Web Text))

Max Activating Token Index: 316

Text #17

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.2985. Min Act: -0.0000

Data Index: 936754 (C4 (Web Text))

Max Activating Token Index: 785

Text #18

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.3415. Min Act: -0.0000

Data Index: 97582 (C4 (Web Text))

Max Activating Token Index: 621

Text #19

Max Range: 2.5591. Min Range: -2.5591

Max Act: 1.2493. Min Act: -0.0000

Data Index: 696950 (C4 (Web Text))

Max Activating Token Index: 761