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

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

Neuron 5066 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5567. Min Act: -0.0001

Data Index: 440833 (C4 (Web Text))

Max Activating Token Index: 301

Text #1

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5284. Min Act: -0.0000

Data Index: 484889 (C4 (Web Text))

Max Activating Token Index: 234

Text #2

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5531. Min Act: -0.0001

Data Index: 443578 (C4 (Web Text))

Max Activating Token Index: 582

Text #3

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5419. Min Act: -0.0000

Data Index: 1340387 (C4 (Web Text))

Max Activating Token Index: 7

Text #4

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5375. Min Act: -0.0000

Data Index: 987949 (C4 (Web Text))

Max Activating Token Index: 677

Text #5

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4626. Min Act: -0.0001

Data Index: 173802 (C4 (Web Text))

Max Activating Token Index: 254

Text #6

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5139. Min Act: -0.0000

Data Index: 166523 (C4 (Web Text))

Max Activating Token Index: 438

Text #7

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4809. Min Act: -0.0001

Data Index: 322569 (C4 (Web Text))

Max Activating Token Index: 407

Text #8

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5273. Min Act: -0.0001

Data Index: 330272 (C4 (Web Text))

Max Activating Token Index: 133

Text #9

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5010. Min Act: -0.0000

Data Index: 537439 (C4 (Web Text))

Max Activating Token Index: 24

Text #10

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.5073. Min Act: -0.0000

Data Index: 478703 (C4 (Web Text))

Max Activating Token Index: 437

Text #11

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4687. Min Act: -0.0001

Data Index: 928748 (C4 (Web Text))

Max Activating Token Index: 804

Text #12

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4967. Min Act: -0.0001

Data Index: 95287 (C4 (Web Text))

Max Activating Token Index: 179

Text #13

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4439. Min Act: -0.0000

Data Index: 1105072 (C4 (Web Text))

Max Activating Token Index: 660

Text #14

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4149. Min Act: -0.0000

Data Index: 674166 (C4 (Web Text))

Max Activating Token Index: 653

Text #15

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4076. Min Act: -0.0000

Data Index: 85040 (C4 (Web Text))

Max Activating Token Index: 123

Text #16

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4574. Min Act: -0.0001

Data Index: 1123258 (C4 (Web Text))

Max Activating Token Index: 641

Text #17

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4325. Min Act: -0.0000

Data Index: 1219651 (C4 (Web Text))

Max Activating Token Index: 15

Text #18

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4691. Min Act: -0.0001

Data Index: 165939 (C4 (Web Text))

Max Activating Token Index: 645

Text #19

Max Range: 0.5567. Min Range: -0.5567

Max Act: 0.4565. Min Act: -0.0001

Data Index: 528167 (C4 (Web Text))

Max Activating Token Index: 266