Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1103 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0105. Min Act: -0.0006

Data Index: 410512 (C4 (Web Text))

Max Activating Token Index: 752

Text #1

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0102. Min Act: -0.0006

Data Index: 359692 (C4 (Web Text))

Max Activating Token Index: 669

Text #2

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0103. Min Act: -0.0005

Data Index: 1378926 (Python Code)

Max Activating Token Index: 288

Text #3

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0103. Min Act: -0.0006

Data Index: 793037 (C4 (Web Text))

Max Activating Token Index: 595

Text #4

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0102. Min Act: -0.0005

Data Index: 1643779 (Python Code)

Max Activating Token Index: 646

Text #5

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0099. Min Act: -0.0005

Data Index: 1426873 (Python Code)

Max Activating Token Index: 857

Text #6

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0101. Min Act: -0.0005

Data Index: 1637134 (Python Code)

Max Activating Token Index: 72

Text #7

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0100. Min Act: -0.0005

Data Index: 283974 (C4 (Web Text))

Max Activating Token Index: 1010

Text #8

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0098. Min Act: -0.0006

Data Index: 276674 (C4 (Web Text))

Max Activating Token Index: 239

Text #9

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0098. Min Act: -0.0005

Data Index: 90995 (C4 (Web Text))

Max Activating Token Index: 809

Text #10

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0097. Min Act: -0.0006

Data Index: 542142 (C4 (Web Text))

Max Activating Token Index: 7

Text #11

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0095. Min Act: -0.0006

Data Index: 47523 (C4 (Web Text))

Max Activating Token Index: 42

Text #12

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0094. Min Act: -0.0006

Data Index: 866698 (C4 (Web Text))

Max Activating Token Index: 357

Text #13

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0097. Min Act: -0.0006

Data Index: 976441 (C4 (Web Text))

Max Activating Token Index: 307

Text #14

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0094. Min Act: -0.0006

Data Index: 450133 (C4 (Web Text))

Max Activating Token Index: 660

Text #15

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0094. Min Act: -0.0006

Data Index: 1301816 (C4 (Web Text))

Max Activating Token Index: 249

Text #16

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0091. Min Act: -0.0006

Data Index: 839670 (C4 (Web Text))

Max Activating Token Index: 820

Text #17

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0094. Min Act: -0.0005

Data Index: 1626131 (Python Code)

Max Activating Token Index: 194

Text #18

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0093. Min Act: -0.0006

Data Index: 153930 (C4 (Web Text))

Max Activating Token Index: 290

Text #19

Max Range: 0.0105. Min Range: -0.0105

Max Act: 0.0092. Min Act: -0.0006

Data Index: 1055195 (C4 (Web Text))

Max Activating Token Index: 4