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

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

Neuron 4785 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.6402. Min Act: -0.0001

Data Index: 1878 (C4 (Web Text))

Max Activating Token Index: 960

Text #1

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.6405. Min Act: -0.0001

Data Index: 1232278 (C4 (Web Text))

Max Activating Token Index: 462

Text #2

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.6421. Min Act: -0.0001

Data Index: 1141500 (C4 (Web Text))

Max Activating Token Index: 371

Text #3

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.6493. Min Act: -0.0001

Data Index: 221731 (C4 (Web Text))

Max Activating Token Index: 729

Text #4

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.6422. Min Act: -0.0001

Data Index: 1003173 (C4 (Web Text))

Max Activating Token Index: 611

Text #5

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.5640. Min Act: -0.0001

Data Index: 1194884 (C4 (Web Text))

Max Activating Token Index: 154

Text #6

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.5026. Min Act: -0.0001

Data Index: 239862 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.5143. Min Act: -0.0001

Data Index: 45151 (C4 (Web Text))

Max Activating Token Index: 26

Text #8

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4786. Min Act: -0.0001

Data Index: 896220 (C4 (Web Text))

Max Activating Token Index: 501

Text #9

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4989. Min Act: -0.0001

Data Index: 190769 (C4 (Web Text))

Max Activating Token Index: 110

Text #10

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4853. Min Act: -0.0001

Data Index: 789985 (C4 (Web Text))

Max Activating Token Index: 958

Text #11

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4942. Min Act: -0.0001

Data Index: 1271011 (C4 (Web Text))

Max Activating Token Index: 705

Text #12

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4948. Min Act: -0.0001

Data Index: 851661 (C4 (Web Text))

Max Activating Token Index: 215

Text #13

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4929. Min Act: -0.0001

Data Index: 1005603 (C4 (Web Text))

Max Activating Token Index: 634

Text #14

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4743. Min Act: -0.0001

Data Index: 323353 (C4 (Web Text))

Max Activating Token Index: 7

Text #15

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.5041. Min Act: -0.0001

Data Index: 319985 (C4 (Web Text))

Max Activating Token Index: 625

Text #16

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4753. Min Act: -0.0001

Data Index: 601293 (C4 (Web Text))

Max Activating Token Index: 894

Text #17

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4750. Min Act: -0.0001

Data Index: 962700 (C4 (Web Text))

Max Activating Token Index: 980

Text #18

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4282. Min Act: -0.0001

Data Index: 906926 (C4 (Web Text))

Max Activating Token Index: 722

Text #19

Max Range: 0.6493. Min Range: -0.6493

Max Act: 0.4617. Min Act: -0.0001

Data Index: 682757 (C4 (Web Text))

Max Activating Token Index: 3