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

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

Neuron 2477 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6955. Min Act: -0.0002

Data Index: 1359103 (C4 (Web Text))

Max Activating Token Index: 66

Text #1

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7948. Min Act: -0.0002

Data Index: 558450 (C4 (Web Text))

Max Activating Token Index: 531

Text #2

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7307. Min Act: -0.0001

Data Index: 1464646 (Python Code)

Max Activating Token Index: 982

Text #3

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7979. Min Act: -0.0001

Data Index: 1417658 (Python Code)

Max Activating Token Index: 982

Text #4

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7515. Min Act: -0.0002

Data Index: 1323310 (C4 (Web Text))

Max Activating Token Index: 905

Text #5

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7884. Min Act: -0.0001

Data Index: 89174 (C4 (Web Text))

Max Activating Token Index: 348

Text #6

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7155. Min Act: -0.0001

Data Index: 538897 (C4 (Web Text))

Max Activating Token Index: 224

Text #7

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7518. Min Act: -0.0001

Data Index: 118811 (C4 (Web Text))

Max Activating Token Index: 657

Text #8

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6406. Min Act: -0.0001

Data Index: 656028 (C4 (Web Text))

Max Activating Token Index: 856

Text #9

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6848. Min Act: -0.0001

Data Index: 1501382 (Python Code)

Max Activating Token Index: 816

Text #10

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6599. Min Act: -0.0001

Data Index: 612104 (C4 (Web Text))

Max Activating Token Index: 279

Text #11

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6586. Min Act: -0.0001

Data Index: 1560070 (Python Code)

Max Activating Token Index: 993

Text #12

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6938. Min Act: -0.0002

Data Index: 1012945 (C4 (Web Text))

Max Activating Token Index: 781

Text #13

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6488. Min Act: -0.0002

Data Index: 667156 (C4 (Web Text))

Max Activating Token Index: 136

Text #14

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6672. Min Act: -0.0001

Data Index: 602680 (C4 (Web Text))

Max Activating Token Index: 395

Text #15

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6430. Min Act: -0.0001

Data Index: 1574023 (Python Code)

Max Activating Token Index: 533

Text #16

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.7327. Min Act: -0.0002

Data Index: 658647 (C4 (Web Text))

Max Activating Token Index: 706

Text #17

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6369. Min Act: -0.0001

Data Index: 1619422 (Python Code)

Max Activating Token Index: 245

Text #18

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.6335. Min Act: -0.0002

Data Index: 1124094 (C4 (Web Text))

Max Activating Token Index: 849

Text #19

Max Range: 0.7979. Min Range: -0.7979

Max Act: 0.5827. Min Act: -0.0001

Data Index: 1286371 (C4 (Web Text))

Max Activating Token Index: 567