Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 45 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.0853. Min Range: -1.0853

Max Act: 1.0853. Min Act: -0.0001

Data Index: 20817 (C4 (Web Text))

Max Activating Token Index: 230

Text #1

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.8505. Min Act: -0.0001

Data Index: 988455 (C4 (Web Text))

Max Activating Token Index: 804

Text #2

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.7831. Min Act: -0.0001

Data Index: 1335514 (C4 (Web Text))

Max Activating Token Index: 972

Text #3

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6532. Min Act: -0.0001

Data Index: 992059 (C4 (Web Text))

Max Activating Token Index: 193

Text #4

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6945. Min Act: -0.0001

Data Index: 499291 (C4 (Web Text))

Max Activating Token Index: 743

Text #5

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.7004. Min Act: -0.0001

Data Index: 650782 (C4 (Web Text))

Max Activating Token Index: 911

Text #6

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.7107. Min Act: -0.0001

Data Index: 267095 (C4 (Web Text))

Max Activating Token Index: 29

Text #7

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.7164. Min Act: -0.0001

Data Index: 1134950 (C4 (Web Text))

Max Activating Token Index: 1010

Text #8

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6946. Min Act: -0.0001

Data Index: 784429 (C4 (Web Text))

Max Activating Token Index: 388

Text #9

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6762. Min Act: -0.0001

Data Index: 566945 (C4 (Web Text))

Max Activating Token Index: 741

Text #10

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.7656. Min Act: -0.0001

Data Index: 571356 (C4 (Web Text))

Max Activating Token Index: 268

Text #11

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6921. Min Act: -0.0001

Data Index: 1251873 (C4 (Web Text))

Max Activating Token Index: 901

Text #12

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6866. Min Act: -0.0001

Data Index: 892920 (C4 (Web Text))

Max Activating Token Index: 624

Text #13

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6739. Min Act: -0.0001

Data Index: 813521 (C4 (Web Text))

Max Activating Token Index: 844

Text #14

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6254. Min Act: -0.0001

Data Index: 1346062 (C4 (Web Text))

Max Activating Token Index: 679

Text #15

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6432. Min Act: -0.0001

Data Index: 1326169 (C4 (Web Text))

Max Activating Token Index: 219

Text #16

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6693. Min Act: -0.0001

Data Index: 413326 (C4 (Web Text))

Max Activating Token Index: 55

Text #17

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6700. Min Act: -0.0001

Data Index: 394456 (C4 (Web Text))

Max Activating Token Index: 150

Text #18

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6732. Min Act: -0.0001

Data Index: 837032 (C4 (Web Text))

Max Activating Token Index: 104

Text #19

Max Range: 1.0853. Min Range: -1.0853

Max Act: 0.6047. Min Act: -0.0001

Data Index: 1604789 (Python Code)

Max Activating Token Index: 783