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

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

Neuron 314 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0839. Min Range: -1.0839

Max Act: 1.0149. Min Act: -0.0001

Data Index: 894303 (C4 (Web Text))

Max Activating Token Index: 76

Text #1

Max Range: 1.0839. Min Range: -1.0839

Max Act: 1.0839. Min Act: -0.0001

Data Index: 505502 (C4 (Web Text))

Max Activating Token Index: 591

Text #2

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.8659. Min Act: -0.0001

Data Index: 1466538 (Python Code)

Max Activating Token Index: 880

Text #3

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.8445. Min Act: -0.0001

Data Index: 1209788 (C4 (Web Text))

Max Activating Token Index: 724

Text #4

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7953. Min Act: -0.0001

Data Index: 1469075 (Python Code)

Max Activating Token Index: 902

Text #5

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7404. Min Act: -0.0001

Data Index: 1168663 (C4 (Web Text))

Max Activating Token Index: 31

Text #6

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7886. Min Act: -0.0001

Data Index: 463215 (C4 (Web Text))

Max Activating Token Index: 483

Text #7

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.8155. Min Act: -0.0001

Data Index: 415559 (C4 (Web Text))

Max Activating Token Index: 575

Text #8

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6854. Min Act: -0.0001

Data Index: 525021 (C4 (Web Text))

Max Activating Token Index: 496

Text #9

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7496. Min Act: -0.0001

Data Index: 319524 (C4 (Web Text))

Max Activating Token Index: 798

Text #10

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6983. Min Act: -0.0001

Data Index: 41528 (C4 (Web Text))

Max Activating Token Index: 779

Text #11

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6890. Min Act: -0.0001

Data Index: 882607 (C4 (Web Text))

Max Activating Token Index: 609

Text #12

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7622. Min Act: -0.0001

Data Index: 567259 (C4 (Web Text))

Max Activating Token Index: 250

Text #13

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7519. Min Act: -0.0001

Data Index: 1317326 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6852. Min Act: -0.0001

Data Index: 854860 (C4 (Web Text))

Max Activating Token Index: 335

Text #15

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.5661. Min Act: -0.0001

Data Index: 651966 (C4 (Web Text))

Max Activating Token Index: 856

Text #16

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7027. Min Act: -0.0001

Data Index: 1514912 (Python Code)

Max Activating Token Index: 220

Text #17

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6303. Min Act: -0.0001

Data Index: 711599 (C4 (Web Text))

Max Activating Token Index: 609

Text #18

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.6344. Min Act: -0.0001

Data Index: 705701 (C4 (Web Text))

Max Activating Token Index: 170

Text #19

Max Range: 1.0839. Min Range: -1.0839

Max Act: 0.7166. Min Act: -0.0001

Data Index: 832097 (C4 (Web Text))

Max Activating Token Index: 476