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

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

Neuron 70 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: 0.0066. Min Range: -0.0066

Max Act: 0.0066. Min Act: -0.0006

Data Index: 552143 (C4 (Web Text))

Max Activating Token Index: 608

Text #1

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0065. Min Act: -0.0006

Data Index: 405033 (C4 (Web Text))

Max Activating Token Index: 111

Text #2

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0059. Min Act: -0.0006

Data Index: 782662 (C4 (Web Text))

Max Activating Token Index: 430

Text #3

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0060. Min Act: -0.0006

Data Index: 824169 (C4 (Web Text))

Max Activating Token Index: 747

Text #4

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0059. Min Act: -0.0006

Data Index: 195329 (C4 (Web Text))

Max Activating Token Index: 365

Text #5

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0058. Min Act: -0.0006

Data Index: 70607 (C4 (Web Text))

Max Activating Token Index: 220

Text #6

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0058. Min Act: -0.0006

Data Index: 165908 (C4 (Web Text))

Max Activating Token Index: 955

Text #7

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0060. Min Act: -0.0006

Data Index: 1322383 (C4 (Web Text))

Max Activating Token Index: 587

Text #8

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0059. Min Act: -0.0006

Data Index: 673800 (C4 (Web Text))

Max Activating Token Index: 79

Text #9

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0057. Min Act: -0.0006

Data Index: 250841 (C4 (Web Text))

Max Activating Token Index: 148

Text #10

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0058. Min Act: -0.0006

Data Index: 1289849 (C4 (Web Text))

Max Activating Token Index: 805

Text #11

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0057. Min Act: -0.0006

Data Index: 1309566 (C4 (Web Text))

Max Activating Token Index: 763

Text #12

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0058. Min Act: -0.0006

Data Index: 480450 (C4 (Web Text))

Max Activating Token Index: 163

Text #13

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0057. Min Act: -0.0006

Data Index: 964955 (C4 (Web Text))

Max Activating Token Index: 826

Text #14

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0058. Min Act: -0.0006

Data Index: 125221 (C4 (Web Text))

Max Activating Token Index: 770

Text #15

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0057. Min Act: -0.0006

Data Index: 794656 (C4 (Web Text))

Max Activating Token Index: 916

Text #16

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0056. Min Act: -0.0006

Data Index: 748987 (C4 (Web Text))

Max Activating Token Index: 614

Text #17

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0055. Min Act: -0.0006

Data Index: 1569781 (Python Code)

Max Activating Token Index: 167

Text #18

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0056. Min Act: -0.0006

Data Index: 877681 (C4 (Web Text))

Max Activating Token Index: 30

Text #19

Max Range: 0.0066. Min Range: -0.0066

Max Act: 0.0056. Min Act: -0.0006

Data Index: 203042 (C4 (Web Text))

Max Activating Token Index: 223