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

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

Neuron 296 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: 0.1276. Min Range: -0.1276

Max Act: 0.1276. Min Act: -0.0000

Data Index: 394669 (C4 (Web Text))

Max Activating Token Index: 625

Text #1

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1206. Min Act: -0.0000

Data Index: 648381 (C4 (Web Text))

Max Activating Token Index: 124

Text #2

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1222. Min Act: -0.0000

Data Index: 840067 (C4 (Web Text))

Max Activating Token Index: 54

Text #3

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1123. Min Act: -0.0001

Data Index: 26127 (C4 (Web Text))

Max Activating Token Index: 310

Text #4

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1066. Min Act: -0.0000

Data Index: 983367 (C4 (Web Text))

Max Activating Token Index: 838

Text #5

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1106. Min Act: -0.0000

Data Index: 158780 (C4 (Web Text))

Max Activating Token Index: 686

Text #6

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1089. Min Act: -0.0001

Data Index: 1214212 (C4 (Web Text))

Max Activating Token Index: 280

Text #7

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1010. Min Act: -0.0000

Data Index: 190 (C4 (Web Text))

Max Activating Token Index: 560

Text #8

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1004. Min Act: -0.0000

Data Index: 936679 (C4 (Web Text))

Max Activating Token Index: 268

Text #9

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.1033. Min Act: -0.0000

Data Index: 788583 (C4 (Web Text))

Max Activating Token Index: 378

Text #10

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0938. Min Act: -0.0001

Data Index: 287060 (C4 (Web Text))

Max Activating Token Index: 559

Text #11

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0898. Min Act: -0.0001

Data Index: 645434 (C4 (Web Text))

Max Activating Token Index: 734

Text #12

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0922. Min Act: -0.0001

Data Index: 995630 (C4 (Web Text))

Max Activating Token Index: 62

Text #13

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0987. Min Act: -0.0000

Data Index: 138046 (C4 (Web Text))

Max Activating Token Index: 849

Text #14

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0951. Min Act: -0.0000

Data Index: 73863 (C4 (Web Text))

Max Activating Token Index: 923

Text #15

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0882. Min Act: -0.0001

Data Index: 1367791 (Python Code)

Max Activating Token Index: 198

Text #16

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0920. Min Act: -0.0000

Data Index: 349907 (C4 (Web Text))

Max Activating Token Index: 268

Text #17

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0864. Min Act: -0.0000

Data Index: 1287741 (C4 (Web Text))

Max Activating Token Index: 967

Text #18

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0923. Min Act: -0.0001

Data Index: 299626 (C4 (Web Text))

Max Activating Token Index: 202

Text #19

Max Range: 0.1276. Min Range: -0.1276

Max Act: 0.0881. Min Act: -0.0001

Data Index: 13117 (C4 (Web Text))

Max Activating Token Index: 534