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

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

Neuron 3770 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.4935. Min Act: -0.0000

Data Index: 225575 (C4 (Web Text))

Max Activating Token Index: 332

Text #1

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.2931. Min Act: -0.0000

Data Index: 219413 (C4 (Web Text))

Max Activating Token Index: 141

Text #2

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.3228. Min Act: -0.0000

Data Index: 273162 (C4 (Web Text))

Max Activating Token Index: 131

Text #3

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1929. Min Act: -0.0000

Data Index: 1433383 (Python Code)

Max Activating Token Index: 735

Text #4

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1626. Min Act: -0.0000

Data Index: 1617897 (Python Code)

Max Activating Token Index: 591

Text #5

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1518. Min Act: -0.0001

Data Index: 966049 (C4 (Web Text))

Max Activating Token Index: 243

Text #6

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1508. Min Act: -0.0001

Data Index: 1600995 (Python Code)

Max Activating Token Index: 762

Text #7

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.2242. Min Act: -0.0001

Data Index: 795520 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1465. Min Act: -0.0000

Data Index: 1102495 (C4 (Web Text))

Max Activating Token Index: 619

Text #9

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1109. Min Act: -0.0000

Data Index: 1016153 (C4 (Web Text))

Max Activating Token Index: 464

Text #10

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1677. Min Act: -0.0000

Data Index: 1245805 (C4 (Web Text))

Max Activating Token Index: 204

Text #11

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1374. Min Act: -0.0000

Data Index: 1464253 (Python Code)

Max Activating Token Index: 473

Text #12

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1248. Min Act: -0.0000

Data Index: 1425429 (Python Code)

Max Activating Token Index: 690

Text #13

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1435. Min Act: -0.0000

Data Index: 799077 (C4 (Web Text))

Max Activating Token Index: 442

Text #14

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1291. Min Act: -0.0000

Data Index: 1581111 (Python Code)

Max Activating Token Index: 728

Text #15

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.0874. Min Act: -0.0001

Data Index: 855774 (C4 (Web Text))

Max Activating Token Index: 887

Text #16

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1220. Min Act: -0.0000

Data Index: 1576721 (Python Code)

Max Activating Token Index: 341

Text #17

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1238. Min Act: -0.0000

Data Index: 323275 (C4 (Web Text))

Max Activating Token Index: 833

Text #18

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1200. Min Act: -0.0001

Data Index: 1439349 (Python Code)

Max Activating Token Index: 166

Text #19

Max Range: 0.4935. Min Range: -0.4935

Max Act: 0.1267. Min Act: -0.0000

Data Index: 1069493 (C4 (Web Text))

Max Activating Token Index: 698