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

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

Neuron 1340 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.1050. Min Act: -0.0001

Data Index: 1159290 (C4 (Web Text))

Max Activating Token Index: 462

Text #1

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0958. Min Act: -0.0001

Data Index: 1471729 (Python Code)

Max Activating Token Index: 15

Text #2

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0826. Min Act: -0.0001

Data Index: 1017930 (C4 (Web Text))

Max Activating Token Index: 786

Text #3

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0813. Min Act: -0.0001

Data Index: 534565 (C4 (Web Text))

Max Activating Token Index: 821

Text #4

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0832. Min Act: -0.0001

Data Index: 1396630 (Python Code)

Max Activating Token Index: 832

Text #5

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0768. Min Act: -0.0001

Data Index: 502278 (C4 (Web Text))

Max Activating Token Index: 148

Text #6

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0713. Min Act: -0.0001

Data Index: 1563943 (Python Code)

Max Activating Token Index: 303

Text #7

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0665. Min Act: -0.0001

Data Index: 910013 (C4 (Web Text))

Max Activating Token Index: 850

Text #8

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0733. Min Act: -0.0001

Data Index: 966437 (C4 (Web Text))

Max Activating Token Index: 845

Text #9

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0671. Min Act: -0.0001

Data Index: 1172406 (C4 (Web Text))

Max Activating Token Index: 612

Text #10

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0671. Min Act: -0.0001

Data Index: 1131423 (C4 (Web Text))

Max Activating Token Index: 69

Text #11

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0693. Min Act: -0.0001

Data Index: 1049864 (C4 (Web Text))

Max Activating Token Index: 107

Text #12

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0566. Min Act: -0.0001

Data Index: 718905 (C4 (Web Text))

Max Activating Token Index: 891

Text #13

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0689. Min Act: -0.0001

Data Index: 1564670 (Python Code)

Max Activating Token Index: 846

Text #14

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0664. Min Act: -0.0001

Data Index: 35926 (C4 (Web Text))

Max Activating Token Index: 965

Text #15

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0670. Min Act: -0.0001

Data Index: 651125 (C4 (Web Text))

Max Activating Token Index: 233

Text #16

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0664. Min Act: -0.0001

Data Index: 578852 (C4 (Web Text))

Max Activating Token Index: 565

Text #17

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0706. Min Act: -0.0001

Data Index: 1041370 (C4 (Web Text))

Max Activating Token Index: 466

Text #18

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0696. Min Act: -0.0001

Data Index: 1358346 (C4 (Web Text))

Max Activating Token Index: 261

Text #19

Max Range: 0.1050. Min Range: -0.1050

Max Act: 0.0658. Min Act: -0.0001

Data Index: 146164 (C4 (Web Text))

Max Activating Token Index: 87