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

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

Neuron 684 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8905. Min Act: -0.0001

Data Index: 1462572 (Python Code)

Max Activating Token Index: 836

Text #1

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8834. Min Act: -0.0001

Data Index: 890508 (C4 (Web Text))

Max Activating Token Index: 113

Text #2

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.9368. Min Act: -0.0001

Data Index: 1521945 (Python Code)

Max Activating Token Index: 11

Text #3

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8964. Min Act: -0.0001

Data Index: 1525371 (Python Code)

Max Activating Token Index: 14

Text #4

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8845. Min Act: -0.0001

Data Index: 1435734 (Python Code)

Max Activating Token Index: 15

Text #5

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8121. Min Act: -0.0001

Data Index: 93272 (C4 (Web Text))

Max Activating Token Index: 91

Text #6

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8152. Min Act: -0.0001

Data Index: 858845 (C4 (Web Text))

Max Activating Token Index: 363

Text #7

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8372. Min Act: -0.0001

Data Index: 1522181 (Python Code)

Max Activating Token Index: 22

Text #8

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8337. Min Act: -0.0001

Data Index: 1575391 (Python Code)

Max Activating Token Index: 992

Text #9

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8079. Min Act: -0.0001

Data Index: 1622880 (Python Code)

Max Activating Token Index: 862

Text #10

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8488. Min Act: -0.0001

Data Index: 1363179 (Python Code)

Max Activating Token Index: 353

Text #11

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8536. Min Act: -0.0001

Data Index: 1573286 (Python Code)

Max Activating Token Index: 890

Text #12

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8517. Min Act: -0.0001

Data Index: 1624285 (Python Code)

Max Activating Token Index: 11

Text #13

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7801. Min Act: -0.0001

Data Index: 1389884 (Python Code)

Max Activating Token Index: 47

Text #14

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7691. Min Act: -0.0001

Data Index: 1575387 (Python Code)

Max Activating Token Index: 33

Text #15

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7942. Min Act: -0.0001

Data Index: 1391347 (Python Code)

Max Activating Token Index: 208

Text #16

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7552. Min Act: -0.0001

Data Index: 1475634 (Python Code)

Max Activating Token Index: 24

Text #17

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.8528. Min Act: -0.0001

Data Index: 1622665 (Python Code)

Max Activating Token Index: 539

Text #18

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7546. Min Act: -0.0001

Data Index: 1603134 (Python Code)

Max Activating Token Index: 810

Text #19

Max Range: 0.9368. Min Range: -0.9368

Max Act: 0.7759. Min Act: -0.0001

Data Index: 1424379 (Python Code)

Max Activating Token Index: 4