Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3822 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.5168. Min Act: -0.0000

Data Index: 774353 (C4 (Web Text))

Max Activating Token Index: 247

Text #1

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.5076. Min Act: -0.0000

Data Index: 683624 (C4 (Web Text))

Max Activating Token Index: 43

Text #2

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4862. Min Act: -0.0001

Data Index: 1109754 (C4 (Web Text))

Max Activating Token Index: 26

Text #3

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4573. Min Act: -0.0001

Data Index: 986496 (C4 (Web Text))

Max Activating Token Index: 25

Text #4

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4591. Min Act: -0.0001

Data Index: 939180 (C4 (Web Text))

Max Activating Token Index: 384

Text #5

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4405. Min Act: -0.0001

Data Index: 773833 (C4 (Web Text))

Max Activating Token Index: 599

Text #6

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4206. Min Act: -0.0001

Data Index: 279397 (C4 (Web Text))

Max Activating Token Index: 499

Text #7

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3963. Min Act: -0.0001

Data Index: 979993 (C4 (Web Text))

Max Activating Token Index: 165

Text #8

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4270. Min Act: -0.0000

Data Index: 640328 (C4 (Web Text))

Max Activating Token Index: 733

Text #9

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4053. Min Act: -0.0000

Data Index: 849926 (C4 (Web Text))

Max Activating Token Index: 272

Text #10

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3915. Min Act: -0.0001

Data Index: 703867 (C4 (Web Text))

Max Activating Token Index: 922

Text #11

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3968. Min Act: -0.0001

Data Index: 178802 (C4 (Web Text))

Max Activating Token Index: 896

Text #12

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.4388. Min Act: -0.0001

Data Index: 50840 (C4 (Web Text))

Max Activating Token Index: 780

Text #13

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3913. Min Act: -0.0000

Data Index: 334041 (C4 (Web Text))

Max Activating Token Index: 70

Text #14

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3897. Min Act: -0.0001

Data Index: 416421 (C4 (Web Text))

Max Activating Token Index: 59

Text #15

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3900. Min Act: -0.0001

Data Index: 248881 (C4 (Web Text))

Max Activating Token Index: 883

Text #16

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3759. Min Act: -0.0001

Data Index: 963063 (C4 (Web Text))

Max Activating Token Index: 85

Text #17

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3732. Min Act: -0.0001

Data Index: 565749 (C4 (Web Text))

Max Activating Token Index: 360

Text #18

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3695. Min Act: -0.0000

Data Index: 122622 (C4 (Web Text))

Max Activating Token Index: 517

Text #19

Max Range: 0.5168. Min Range: -0.5168

Max Act: 0.3634. Min Act: -0.0001

Data Index: 1301113 (C4 (Web Text))

Max Activating Token Index: 71