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

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

Neuron 1708 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8886. Min Act: -0.0003

Data Index: 670141 (C4 (Web Text))

Max Activating Token Index: 152

Text #1

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8761. Min Act: -0.0003

Data Index: 1282101 (C4 (Web Text))

Max Activating Token Index: 14

Text #2

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8408. Min Act: -0.0003

Data Index: 940735 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8263. Min Act: -0.0003

Data Index: 1281157 (C4 (Web Text))

Max Activating Token Index: 761

Text #4

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8171. Min Act: -0.0003

Data Index: 929174 (C4 (Web Text))

Max Activating Token Index: 62

Text #5

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7934. Min Act: -0.0003

Data Index: 844686 (C4 (Web Text))

Max Activating Token Index: 299

Text #6

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8892. Min Act: -0.0003

Data Index: 1146686 (C4 (Web Text))

Max Activating Token Index: 353

Text #7

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8436. Min Act: -0.0003

Data Index: 696886 (C4 (Web Text))

Max Activating Token Index: 30

Text #8

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8016. Min Act: -0.0003

Data Index: 1134767 (C4 (Web Text))

Max Activating Token Index: 1003

Text #9

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.8287. Min Act: -0.0003

Data Index: 1252555 (C4 (Web Text))

Max Activating Token Index: 645

Text #10

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7794. Min Act: -0.0003

Data Index: 1143598 (C4 (Web Text))

Max Activating Token Index: 847

Text #11

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7758. Min Act: -0.0003

Data Index: 139916 (C4 (Web Text))

Max Activating Token Index: 45

Text #12

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7771. Min Act: -0.0003

Data Index: 637336 (C4 (Web Text))

Max Activating Token Index: 294

Text #13

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7724. Min Act: -0.0003

Data Index: 398789 (C4 (Web Text))

Max Activating Token Index: 41

Text #14

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7395. Min Act: -0.0003

Data Index: 255371 (C4 (Web Text))

Max Activating Token Index: 1001

Text #15

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7643. Min Act: -0.0003

Data Index: 476568 (C4 (Web Text))

Max Activating Token Index: 49

Text #16

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7385. Min Act: -0.0003

Data Index: 120101 (C4 (Web Text))

Max Activating Token Index: 174

Text #17

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7588. Min Act: -0.0003

Data Index: 655593 (C4 (Web Text))

Max Activating Token Index: 525

Text #18

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7898. Min Act: -0.0003

Data Index: 1135620 (C4 (Web Text))

Max Activating Token Index: 1009

Text #19

Max Range: 0.8892. Min Range: -0.8892

Max Act: 0.7569. Min Act: -0.0003

Data Index: 1215645 (C4 (Web Text))

Max Activating Token Index: 968