Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 6050 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 5.3067. Min Range: -5.3067

Max Act: 5.3067. Min Act: -0.0000

Data Index: 95453 (C4 (Web Text))

Max Activating Token Index: 495

Text #1

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.8279. Min Act: -0.0000

Data Index: 1122261 (C4 (Web Text))

Max Activating Token Index: 229

Text #2

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.8124. Min Act: -0.0000

Data Index: 1002174 (C4 (Web Text))

Max Activating Token Index: 33

Text #3

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.5364. Min Act: -0.0000

Data Index: 1225010 (C4 (Web Text))

Max Activating Token Index: 937

Text #4

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.6428. Min Act: -0.0000

Data Index: 1271012 (C4 (Web Text))

Max Activating Token Index: 428

Text #5

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.0555. Min Act: -0.0000

Data Index: 38833 (C4 (Web Text))

Max Activating Token Index: 930

Text #6

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.3976. Min Act: -0.0000

Data Index: 530376 (C4 (Web Text))

Max Activating Token Index: 373

Text #7

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.5028. Min Act: -0.0000

Data Index: 1179229 (C4 (Web Text))

Max Activating Token Index: 84

Text #8

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.5015. Min Act: -0.0000

Data Index: 798112 (C4 (Web Text))

Max Activating Token Index: 522

Text #9

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.0349. Min Act: -0.0000

Data Index: 165762 (C4 (Web Text))

Max Activating Token Index: 499

Text #10

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.1778. Min Act: -0.0000

Data Index: 453617 (C4 (Web Text))

Max Activating Token Index: 290

Text #11

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.1470. Min Act: -0.0000

Data Index: 667660 (C4 (Web Text))

Max Activating Token Index: 596

Text #12

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.1238. Min Act: -0.0000

Data Index: 1186005 (C4 (Web Text))

Max Activating Token Index: 434

Text #13

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.1066. Min Act: -0.0000

Data Index: 1181517 (C4 (Web Text))

Max Activating Token Index: 452

Text #14

Max Range: 5.3067. Min Range: -5.3067

Max Act: 4.0864. Min Act: -0.0000

Data Index: 151534 (C4 (Web Text))

Max Activating Token Index: 804

Text #15

Max Range: 5.3067. Min Range: -5.3067

Max Act: 3.8139. Min Act: -0.0000

Data Index: 596354 (C4 (Web Text))

Max Activating Token Index: 164

Text #16

Max Range: 5.3067. Min Range: -5.3067

Max Act: 3.9618. Min Act: -0.0000

Data Index: 906523 (C4 (Web Text))

Max Activating Token Index: 60

Text #17

Max Range: 5.3067. Min Range: -5.3067

Max Act: 3.8342. Min Act: -0.0000

Data Index: 368036 (C4 (Web Text))

Max Activating Token Index: 94

Text #18

Max Range: 5.3067. Min Range: -5.3067

Max Act: 3.6562. Min Act: -0.0000

Data Index: 1166421 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 5.3067. Min Range: -5.3067

Max Act: 3.6309. Min Act: -0.0000

Data Index: 1096373 (C4 (Web Text))

Max Activating Token Index: 331