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

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

Neuron 3712 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1333. Min Act: -0.0000

Data Index: 213200 (C4 (Web Text))

Max Activating Token Index: 227

Text #1

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1261. Min Act: -0.0000

Data Index: 843609 (C4 (Web Text))

Max Activating Token Index: 604

Text #2

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1232. Min Act: -0.0000

Data Index: 911086 (C4 (Web Text))

Max Activating Token Index: 361

Text #3

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1224. Min Act: -0.0000

Data Index: 1221042 (C4 (Web Text))

Max Activating Token Index: 861

Text #4

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1256. Min Act: -0.0000

Data Index: 1263033 (C4 (Web Text))

Max Activating Token Index: 209

Text #5

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1166. Min Act: -0.0000

Data Index: 1206722 (C4 (Web Text))

Max Activating Token Index: 394

Text #6

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1072. Min Act: -0.0000

Data Index: 25770 (C4 (Web Text))

Max Activating Token Index: 935

Text #7

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1081. Min Act: -0.0000

Data Index: 740160 (C4 (Web Text))

Max Activating Token Index: 1003

Text #8

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.1053. Min Act: -0.0000

Data Index: 32685 (C4 (Web Text))

Max Activating Token Index: 1000

Text #9

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0906. Min Act: -0.0000

Data Index: 1009144 (C4 (Web Text))

Max Activating Token Index: 910

Text #10

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0898. Min Act: -0.0000

Data Index: 134723 (C4 (Web Text))

Max Activating Token Index: 233

Text #11

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0864. Min Act: -0.0000

Data Index: 302818 (C4 (Web Text))

Max Activating Token Index: 200

Text #12

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0869. Min Act: -0.0000

Data Index: 11914 (C4 (Web Text))

Max Activating Token Index: 567

Text #13

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0854. Min Act: -0.0000

Data Index: 1183913 (C4 (Web Text))

Max Activating Token Index: 843

Text #14

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0800. Min Act: -0.0000

Data Index: 431871 (C4 (Web Text))

Max Activating Token Index: 52

Text #15

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0843. Min Act: -0.0000

Data Index: 784303 (C4 (Web Text))

Max Activating Token Index: 774

Text #16

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0782. Min Act: -0.0000

Data Index: 175572 (C4 (Web Text))

Max Activating Token Index: 334

Text #17

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0819. Min Act: -0.0000

Data Index: 1203093 (C4 (Web Text))

Max Activating Token Index: 82

Text #18

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0760. Min Act: -0.0000

Data Index: 520881 (C4 (Web Text))

Max Activating Token Index: 685

Text #19

Max Range: 0.1333. Min Range: -0.1333

Max Act: 0.0785. Min Act: -0.0000

Data Index: 236361 (C4 (Web Text))

Max Activating Token Index: 861