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

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

Neuron 2109 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.9846. Min Act: -0.0001

Data Index: 1584194 (Python Code)

Max Activating Token Index: 439

Text #1

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.9676. Min Act: -0.0001

Data Index: 484248 (C4 (Web Text))

Max Activating Token Index: 38

Text #2

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7441. Min Act: -0.0001

Data Index: 1163210 (C4 (Web Text))

Max Activating Token Index: 406

Text #3

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6482. Min Act: -0.0001

Data Index: 1224057 (C4 (Web Text))

Max Activating Token Index: 379

Text #4

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7559. Min Act: -0.0001

Data Index: 121132 (C4 (Web Text))

Max Activating Token Index: 942

Text #5

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7023. Min Act: -0.0001

Data Index: 464671 (C4 (Web Text))

Max Activating Token Index: 816

Text #6

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7441. Min Act: -0.0001

Data Index: 1514743 (Python Code)

Max Activating Token Index: 103

Text #7

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7251. Min Act: -0.0001

Data Index: 1470882 (Python Code)

Max Activating Token Index: 978

Text #8

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7037. Min Act: -0.0001

Data Index: 368391 (C4 (Web Text))

Max Activating Token Index: 1012

Text #9

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6952. Min Act: -0.0001

Data Index: 684299 (C4 (Web Text))

Max Activating Token Index: 584

Text #10

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6904. Min Act: -0.0001

Data Index: 406748 (C4 (Web Text))

Max Activating Token Index: 857

Text #11

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6510. Min Act: -0.0001

Data Index: 1642761 (Python Code)

Max Activating Token Index: 878

Text #12

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6577. Min Act: -0.0001

Data Index: 1414039 (Python Code)

Max Activating Token Index: 1021

Text #13

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6411. Min Act: -0.0001

Data Index: 1558227 (Python Code)

Max Activating Token Index: 551

Text #14

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6672. Min Act: -0.0001

Data Index: 1272828 (C4 (Web Text))

Max Activating Token Index: 897

Text #15

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7057. Min Act: -0.0001

Data Index: 1501238 (Python Code)

Max Activating Token Index: 221

Text #16

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7064. Min Act: -0.0001

Data Index: 1170823 (C4 (Web Text))

Max Activating Token Index: 398

Text #17

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7080. Min Act: -0.0001

Data Index: 951350 (C4 (Web Text))

Max Activating Token Index: 111

Text #18

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.7099. Min Act: -0.0001

Data Index: 1045636 (C4 (Web Text))

Max Activating Token Index: 31

Text #19

Max Range: 0.9846. Min Range: -0.9846

Max Act: 0.6585. Min Act: -0.0001

Data Index: 489066 (C4 (Web Text))

Max Activating Token Index: 868