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

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

Neuron 920 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: 1.1062. Min Range: -1.1062

Max Act: 1.0143. Min Act: -0.0001

Data Index: 540363 (C4 (Web Text))

Max Activating Token Index: 157

Text #1

Max Range: 1.1062. Min Range: -1.1062

Max Act: 1.0761. Min Act: -0.0001

Data Index: 872259 (C4 (Web Text))

Max Activating Token Index: 102

Text #2

Max Range: 1.1062. Min Range: -1.1062

Max Act: 1.1062. Min Act: -0.0001

Data Index: 691304 (C4 (Web Text))

Max Activating Token Index: 61

Text #3

Max Range: 1.1062. Min Range: -1.1062

Max Act: 1.0585. Min Act: -0.0001

Data Index: 1318843 (C4 (Web Text))

Max Activating Token Index: 682

Text #4

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9193. Min Act: -0.0001

Data Index: 692424 (C4 (Web Text))

Max Activating Token Index: 609

Text #5

Max Range: 1.1062. Min Range: -1.1062

Max Act: 1.0311. Min Act: -0.0001

Data Index: 1098620 (C4 (Web Text))

Max Activating Token Index: 122

Text #6

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9296. Min Act: -0.0001

Data Index: 1249486 (C4 (Web Text))

Max Activating Token Index: 188

Text #7

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9611. Min Act: -0.0001

Data Index: 1085130 (C4 (Web Text))

Max Activating Token Index: 19

Text #8

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9813. Min Act: -0.0001

Data Index: 1125541 (C4 (Web Text))

Max Activating Token Index: 89

Text #9

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9525. Min Act: -0.0001

Data Index: 627540 (C4 (Web Text))

Max Activating Token Index: 953

Text #10

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9103. Min Act: -0.0001

Data Index: 300607 (C4 (Web Text))

Max Activating Token Index: 9

Text #11

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9454. Min Act: -0.0001

Data Index: 1113775 (C4 (Web Text))

Max Activating Token Index: 863

Text #12

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9560. Min Act: -0.0001

Data Index: 106160 (C4 (Web Text))

Max Activating Token Index: 151

Text #13

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9344. Min Act: -0.0001

Data Index: 377502 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.8731. Min Act: -0.0001

Data Index: 1382343 (Python Code)

Max Activating Token Index: 111

Text #15

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9474. Min Act: -0.0001

Data Index: 522970 (C4 (Web Text))

Max Activating Token Index: 112

Text #16

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.8841. Min Act: -0.0001

Data Index: 407651 (C4 (Web Text))

Max Activating Token Index: 369

Text #17

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9140. Min Act: -0.0001

Data Index: 177328 (C4 (Web Text))

Max Activating Token Index: 36

Text #18

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9196. Min Act: -0.0001

Data Index: 106473 (C4 (Web Text))

Max Activating Token Index: 130

Text #19

Max Range: 1.1062. Min Range: -1.1062

Max Act: 0.9085. Min Act: -0.0001

Data Index: 442106 (C4 (Web Text))

Max Activating Token Index: 512