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

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

Neuron 3815 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.8778. Min Act: -0.0001

Data Index: 937607 (C4 (Web Text))

Max Activating Token Index: 19

Text #1

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.8714. Min Act: -0.0001

Data Index: 87733 (C4 (Web Text))

Max Activating Token Index: 886

Text #2

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.8786. Min Act: -0.0000

Data Index: 1181239 (C4 (Web Text))

Max Activating Token Index: 881

Text #3

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.8112. Min Act: -0.0001

Data Index: 1160932 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.7703. Min Act: -0.0001

Data Index: 1019290 (C4 (Web Text))

Max Activating Token Index: 967

Text #5

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.8054. Min Act: -0.0001

Data Index: 758406 (C4 (Web Text))

Max Activating Token Index: 611

Text #6

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.7736. Min Act: -0.0000

Data Index: 1283477 (C4 (Web Text))

Max Activating Token Index: 60

Text #7

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.7174. Min Act: -0.0001

Data Index: 1347043 (C4 (Web Text))

Max Activating Token Index: 824

Text #8

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.7131. Min Act: -0.0000

Data Index: 574610 (C4 (Web Text))

Max Activating Token Index: 56

Text #9

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.7244. Min Act: -0.0000

Data Index: 1125848 (C4 (Web Text))

Max Activating Token Index: 555

Text #10

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6730. Min Act: -0.0001

Data Index: 439218 (C4 (Web Text))

Max Activating Token Index: 448

Text #11

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6243. Min Act: -0.0001

Data Index: 635484 (C4 (Web Text))

Max Activating Token Index: 787

Text #12

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6866. Min Act: -0.0000

Data Index: 795546 (C4 (Web Text))

Max Activating Token Index: 87

Text #13

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6822. Min Act: -0.0000

Data Index: 1285757 (C4 (Web Text))

Max Activating Token Index: 136

Text #14

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6801. Min Act: -0.0000

Data Index: 565334 (C4 (Web Text))

Max Activating Token Index: 20

Text #15

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6991. Min Act: -0.0000

Data Index: 223680 (C4 (Web Text))

Max Activating Token Index: 547

Text #16

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6546. Min Act: -0.0000

Data Index: 793416 (C4 (Web Text))

Max Activating Token Index: 929

Text #17

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6945. Min Act: -0.0001

Data Index: 680545 (C4 (Web Text))

Max Activating Token Index: 508

Text #18

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6818. Min Act: -0.0000

Data Index: 190845 (C4 (Web Text))

Max Activating Token Index: 909

Text #19

Max Range: 0.8786. Min Range: -0.8786

Max Act: 0.6400. Min Act: -0.0000

Data Index: 791718 (C4 (Web Text))

Max Activating Token Index: 253