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

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

Neuron 2495 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: 0.2699. Min Range: -0.2699

Max Act: 0.2480. Min Act: -0.0001

Data Index: 1221501 (C4 (Web Text))

Max Activating Token Index: 738

Text #1

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2590. Min Act: -0.0001

Data Index: 996669 (C4 (Web Text))

Max Activating Token Index: 760

Text #2

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2699. Min Act: -0.0001

Data Index: 975905 (C4 (Web Text))

Max Activating Token Index: 580

Text #3

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2531. Min Act: -0.0001

Data Index: 130225 (C4 (Web Text))

Max Activating Token Index: 520

Text #4

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2467. Min Act: -0.0001

Data Index: 297607 (C4 (Web Text))

Max Activating Token Index: 651

Text #5

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2026. Min Act: -0.0001

Data Index: 278949 (C4 (Web Text))

Max Activating Token Index: 603

Text #6

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2331. Min Act: -0.0001

Data Index: 421061 (C4 (Web Text))

Max Activating Token Index: 22

Text #7

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.1940. Min Act: -0.0001

Data Index: 1226623 (C4 (Web Text))

Max Activating Token Index: 733

Text #8

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2274. Min Act: -0.0001

Data Index: 1635696 (Python Code)

Max Activating Token Index: 81

Text #9

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2126. Min Act: -0.0001

Data Index: 1348011 (C4 (Web Text))

Max Activating Token Index: 225

Text #10

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2193. Min Act: -0.0001

Data Index: 270320 (C4 (Web Text))

Max Activating Token Index: 7

Text #11

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2184. Min Act: -0.0001

Data Index: 1039074 (C4 (Web Text))

Max Activating Token Index: 238

Text #12

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2160. Min Act: -0.0001

Data Index: 342493 (C4 (Web Text))

Max Activating Token Index: 112

Text #13

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2062. Min Act: -0.0001

Data Index: 1512157 (Python Code)

Max Activating Token Index: 14

Text #14

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.1729. Min Act: -0.0001

Data Index: 285719 (C4 (Web Text))

Max Activating Token Index: 786

Text #15

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2204. Min Act: -0.0001

Data Index: 436495 (C4 (Web Text))

Max Activating Token Index: 13

Text #16

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2034. Min Act: -0.0001

Data Index: 750735 (C4 (Web Text))

Max Activating Token Index: 113

Text #17

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2012. Min Act: -0.0001

Data Index: 389560 (C4 (Web Text))

Max Activating Token Index: 669

Text #18

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2030. Min Act: -0.0001

Data Index: 239178 (C4 (Web Text))

Max Activating Token Index: 445

Text #19

Max Range: 0.2699. Min Range: -0.2699

Max Act: 0.2094. Min Act: -0.0001

Data Index: 342542 (C4 (Web Text))

Max Activating Token Index: 4