Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3122 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 4.2932. Min Range: -4.2932

Max Act: 4.1153. Min Act: -0.0001

Data Index: 623551 (C4 (Web Text))

Max Activating Token Index: 232

Text #1

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.6231. Min Act: -0.0000

Data Index: 17295 (C4 (Web Text))

Max Activating Token Index: 766

Text #2

Max Range: 4.2932. Min Range: -4.2932

Max Act: 4.2932. Min Act: -0.0001

Data Index: 893231 (C4 (Web Text))

Max Activating Token Index: 112

Text #3

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.7981. Min Act: -0.0001

Data Index: 1564013 (Python Code)

Max Activating Token Index: 602

Text #4

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.5679. Min Act: -0.0001

Data Index: 983625 (C4 (Web Text))

Max Activating Token Index: 735

Text #5

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.4417. Min Act: -0.0000

Data Index: 783685 (C4 (Web Text))

Max Activating Token Index: 213

Text #6

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.4170. Min Act: -0.0001

Data Index: 1161844 (C4 (Web Text))

Max Activating Token Index: 371

Text #7

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.5903. Min Act: -0.0001

Data Index: 964144 (C4 (Web Text))

Max Activating Token Index: 950

Text #8

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.5077. Min Act: -0.0000

Data Index: 1567527 (Python Code)

Max Activating Token Index: 403

Text #9

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.1921. Min Act: -0.0001

Data Index: 1490142 (Python Code)

Max Activating Token Index: 410

Text #10

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.3335. Min Act: -0.0001

Data Index: 1370789 (Python Code)

Max Activating Token Index: 429

Text #11

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.1911. Min Act: -0.0001

Data Index: 734953 (C4 (Web Text))

Max Activating Token Index: 694

Text #12

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.1465. Min Act: -0.0001

Data Index: 164300 (C4 (Web Text))

Max Activating Token Index: 327

Text #13

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.0292. Min Act: -0.0001

Data Index: 1206514 (C4 (Web Text))

Max Activating Token Index: 65

Text #14

Max Range: 4.2932. Min Range: -4.2932

Max Act: 2.8556. Min Act: -0.0001

Data Index: 1439956 (Python Code)

Max Activating Token Index: 160

Text #15

Max Range: 4.2932. Min Range: -4.2932

Max Act: 3.1224. Min Act: -0.0001

Data Index: 534260 (C4 (Web Text))

Max Activating Token Index: 148

Text #16

Max Range: 4.2932. Min Range: -4.2932

Max Act: 2.8832. Min Act: -0.0001

Data Index: 1186535 (C4 (Web Text))

Max Activating Token Index: 272

Text #17

Max Range: 4.2932. Min Range: -4.2932

Max Act: 2.8767. Min Act: -0.0001

Data Index: 378081 (C4 (Web Text))

Max Activating Token Index: 205

Text #18

Max Range: 4.2932. Min Range: -4.2932

Max Act: 2.7210. Min Act: -0.0001

Data Index: 484756 (C4 (Web Text))

Max Activating Token Index: 940

Text #19

Max Range: 4.2932. Min Range: -4.2932

Max Act: 2.6979. Min Act: -0.0001

Data Index: 1072718 (C4 (Web Text))

Max Activating Token Index: 177