Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2653 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0052. Min Act: -0.0003

Data Index: 1558832 (Python Code)

Max Activating Token Index: 993

Text #1

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0051. Min Act: -0.0004

Data Index: 686881 (C4 (Web Text))

Max Activating Token Index: 294

Text #2

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0050. Min Act: -0.0003

Data Index: 426010 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0050. Min Act: -0.0004

Data Index: 1122124 (C4 (Web Text))

Max Activating Token Index: 986

Text #4

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0003

Data Index: 743671 (C4 (Web Text))

Max Activating Token Index: 9

Text #5

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0050. Min Act: -0.0003

Data Index: 1228583 (C4 (Web Text))

Max Activating Token Index: 319

Text #6

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0003

Data Index: 1283325 (C4 (Web Text))

Max Activating Token Index: 8

Text #7

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0004

Data Index: 290639 (C4 (Web Text))

Max Activating Token Index: 996

Text #8

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0003

Data Index: 741162 (C4 (Web Text))

Max Activating Token Index: 47

Text #9

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0003

Data Index: 1017845 (C4 (Web Text))

Max Activating Token Index: 885

Text #10

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 1237939 (C4 (Web Text))

Max Activating Token Index: 47

Text #11

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 858163 (C4 (Web Text))

Max Activating Token Index: 12

Text #12

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0049. Min Act: -0.0003

Data Index: 1214947 (C4 (Web Text))

Max Activating Token Index: 208

Text #13

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 180877 (C4 (Web Text))

Max Activating Token Index: 807

Text #14

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 809901 (C4 (Web Text))

Max Activating Token Index: 650

Text #15

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 1466760 (Python Code)

Max Activating Token Index: 1009

Text #16

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 712947 (C4 (Web Text))

Max Activating Token Index: 935

Text #17

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 133270 (C4 (Web Text))

Max Activating Token Index: 59

Text #18

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0047. Min Act: -0.0003

Data Index: 770627 (C4 (Web Text))

Max Activating Token Index: 207

Text #19

Max Range: 0.0052. Min Range: -0.0052

Max Act: 0.0048. Min Act: -0.0003

Data Index: 168758 (C4 (Web Text))

Max Activating Token Index: 28