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

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

Neuron 4901 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.8647. Min Act: -0.0000

Data Index: 615259 (C4 (Web Text))

Max Activating Token Index: 533

Text #1

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.7054. Min Act: -0.0000

Data Index: 1264345 (C4 (Web Text))

Max Activating Token Index: 604

Text #2

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.7068. Min Act: -0.0000

Data Index: 1062213 (C4 (Web Text))

Max Activating Token Index: 785

Text #3

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.6494. Min Act: -0.0000

Data Index: 223852 (C4 (Web Text))

Max Activating Token Index: 169

Text #4

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5590. Min Act: -0.0000

Data Index: 1193298 (C4 (Web Text))

Max Activating Token Index: 787

Text #5

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5955. Min Act: -0.0000

Data Index: 785467 (C4 (Web Text))

Max Activating Token Index: 497

Text #6

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5940. Min Act: -0.0000

Data Index: 427415 (C4 (Web Text))

Max Activating Token Index: 47

Text #7

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5631. Min Act: -0.0000

Data Index: 157219 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5421. Min Act: -0.0000

Data Index: 186494 (C4 (Web Text))

Max Activating Token Index: 472

Text #9

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5536. Min Act: -0.0000

Data Index: 138875 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5514. Min Act: -0.0000

Data Index: 47741 (C4 (Web Text))

Max Activating Token Index: 389

Text #11

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.4621. Min Act: 0.0000

Data Index: 317742 (C4 (Web Text))

Max Activating Token Index: 801

Text #12

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5134. Min Act: -0.0000

Data Index: 1088250 (C4 (Web Text))

Max Activating Token Index: 77

Text #13

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5040. Min Act: -0.0000

Data Index: 1213237 (C4 (Web Text))

Max Activating Token Index: 907

Text #14

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5160. Min Act: -0.0000

Data Index: 1585482 (Python Code)

Max Activating Token Index: 546

Text #15

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5290. Min Act: -0.0000

Data Index: 261058 (C4 (Web Text))

Max Activating Token Index: 950

Text #16

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.4952. Min Act: -0.0000

Data Index: 787075 (C4 (Web Text))

Max Activating Token Index: 14

Text #17

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5083. Min Act: -0.0000

Data Index: 1156455 (C4 (Web Text))

Max Activating Token Index: 474

Text #18

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.5176. Min Act: -0.0000

Data Index: 285089 (C4 (Web Text))

Max Activating Token Index: 958

Text #19

Max Range: 0.8647. Min Range: -0.8647

Max Act: 0.4950. Min Act: -0.0000

Data Index: 724538 (C4 (Web Text))

Max Activating Token Index: 596