Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 264 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.6095. Min Act: -0.0000

Data Index: 820918 (C4 (Web Text))

Max Activating Token Index: 245

Text #1

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.5459. Min Act: -0.0000

Data Index: 371419 (C4 (Web Text))

Max Activating Token Index: 480

Text #2

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.4941. Min Act: -0.0000

Data Index: 430790 (C4 (Web Text))

Max Activating Token Index: 796

Text #3

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3745. Min Act: -0.0000

Data Index: 810283 (C4 (Web Text))

Max Activating Token Index: 809

Text #4

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3841. Min Act: 0.0000

Data Index: 1351850 (C4 (Web Text))

Max Activating Token Index: 719

Text #5

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.4015. Min Act: -0.0000

Data Index: 914655 (C4 (Web Text))

Max Activating Token Index: 940

Text #6

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3594. Min Act: -0.0000

Data Index: 1405071 (Python Code)

Max Activating Token Index: 750

Text #7

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3448. Min Act: -0.0000

Data Index: 1523417 (Python Code)

Max Activating Token Index: 752

Text #8

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3460. Min Act: -0.0000

Data Index: 283381 (C4 (Web Text))

Max Activating Token Index: 981

Text #9

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3551. Min Act: -0.0000

Data Index: 1307031 (C4 (Web Text))

Max Activating Token Index: 850

Text #10

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3080. Min Act: -0.0000

Data Index: 105797 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2869. Min Act: -0.0000

Data Index: 1617098 (Python Code)

Max Activating Token Index: 197

Text #12

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3123. Min Act: -0.0000

Data Index: 1502052 (Python Code)

Max Activating Token Index: 951

Text #13

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3049. Min Act: -0.0000

Data Index: 1008753 (C4 (Web Text))

Max Activating Token Index: 289

Text #14

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2907. Min Act: -0.0000

Data Index: 1632841 (Python Code)

Max Activating Token Index: 71

Text #15

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2986. Min Act: -0.0000

Data Index: 355347 (C4 (Web Text))

Max Activating Token Index: 1003

Text #16

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.3054. Min Act: -0.0000

Data Index: 1575004 (Python Code)

Max Activating Token Index: 166

Text #17

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2883. Min Act: -0.0000

Data Index: 975094 (C4 (Web Text))

Max Activating Token Index: 96

Text #18

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2691. Min Act: -0.0000

Data Index: 1427546 (Python Code)

Max Activating Token Index: 314

Text #19

Max Range: 0.6095. Min Range: -0.6095

Max Act: 0.2970. Min Act: -0.0000

Data Index: 1608991 (Python Code)

Max Activating Token Index: 16