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

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

Neuron 2948 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: 3.8899. Min Range: -3.8899

Max Act: 3.8899. Min Act: -0.0001

Data Index: 1117256 (C4 (Web Text))

Max Activating Token Index: 18

Text #1

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.7088. Min Act: -0.0001

Data Index: 1143823 (C4 (Web Text))

Max Activating Token Index: 258

Text #2

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.5670. Min Act: -0.0001

Data Index: 1014820 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.2011. Min Act: -0.0001

Data Index: 241204 (C4 (Web Text))

Max Activating Token Index: 97

Text #4

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.2571. Min Act: -0.0001

Data Index: 971474 (C4 (Web Text))

Max Activating Token Index: 29

Text #5

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.2284. Min Act: -0.0001

Data Index: 226730 (C4 (Web Text))

Max Activating Token Index: 20

Text #6

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.0368. Min Act: -0.0001

Data Index: 386079 (C4 (Web Text))

Max Activating Token Index: 17

Text #7

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.0876. Min Act: -0.0001

Data Index: 970697 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.9386. Min Act: -0.0001

Data Index: 1197789 (C4 (Web Text))

Max Activating Token Index: 16

Text #9

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.9003. Min Act: -0.0001

Data Index: 984320 (C4 (Web Text))

Max Activating Token Index: 115

Text #10

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.9582. Min Act: -0.0001

Data Index: 546961 (C4 (Web Text))

Max Activating Token Index: 35

Text #11

Max Range: 3.8899. Min Range: -3.8899

Max Act: 3.0746. Min Act: -0.0001

Data Index: 231835 (C4 (Web Text))

Max Activating Token Index: 13

Text #12

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.9659. Min Act: -0.0001

Data Index: 827386 (C4 (Web Text))

Max Activating Token Index: 11

Text #13

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.8120. Min Act: -0.0001

Data Index: 608830 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.7719. Min Act: -0.0001

Data Index: 875493 (C4 (Web Text))

Max Activating Token Index: 12

Text #15

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.8115. Min Act: -0.0001

Data Index: 303346 (C4 (Web Text))

Max Activating Token Index: 202

Text #16

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.7724. Min Act: -0.0001

Data Index: 417351 (C4 (Web Text))

Max Activating Token Index: 24

Text #17

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.7037. Min Act: -0.0001

Data Index: 195626 (C4 (Web Text))

Max Activating Token Index: 10

Text #18

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.7739. Min Act: -0.0001

Data Index: 1068087 (C4 (Web Text))

Max Activating Token Index: 36

Text #19

Max Range: 3.8899. Min Range: -3.8899

Max Act: 2.7653. Min Act: -0.0001

Data Index: 608922 (C4 (Web Text))

Max Activating Token Index: 20