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

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

Neuron 4075 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: 9.9616. Min Range: -9.9616

Max Act: 9.9616. Min Act: -0.0001

Data Index: 1466650 (Python Code)

Max Activating Token Index: 231

Text #1

Max Range: 9.9616. Min Range: -9.9616

Max Act: 9.5588. Min Act: -0.0001

Data Index: 1608532 (Python Code)

Max Activating Token Index: 765

Text #2

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.4841. Min Act: -0.0001

Data Index: 1567641 (Python Code)

Max Activating Token Index: 141

Text #3

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.5525. Min Act: -0.0001

Data Index: 1208767 (C4 (Web Text))

Max Activating Token Index: 273

Text #4

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.5638. Min Act: -0.0001

Data Index: 1469302 (Python Code)

Max Activating Token Index: 736

Text #5

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.5978. Min Act: -0.0001

Data Index: 709620 (C4 (Web Text))

Max Activating Token Index: 206

Text #6

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.2409. Min Act: -0.0000

Data Index: 1006491 (C4 (Web Text))

Max Activating Token Index: 190

Text #7

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.0226. Min Act: -0.0000

Data Index: 542733 (C4 (Web Text))

Max Activating Token Index: 122

Text #8

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.2376. Min Act: -0.0001

Data Index: 1332333 (C4 (Web Text))

Max Activating Token Index: 73

Text #9

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.1409. Min Act: -0.0001

Data Index: 649919 (C4 (Web Text))

Max Activating Token Index: 386

Text #10

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.4930. Min Act: -0.0001

Data Index: 1633457 (Python Code)

Max Activating Token Index: 591

Text #11

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.3938. Min Act: -0.0001

Data Index: 582146 (C4 (Web Text))

Max Activating Token Index: 890

Text #12

Max Range: 9.9616. Min Range: -9.9616

Max Act: 7.7064. Min Act: -0.0001

Data Index: 977242 (C4 (Web Text))

Max Activating Token Index: 348

Text #13

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.0201. Min Act: -0.0001

Data Index: 817204 (C4 (Web Text))

Max Activating Token Index: 134

Text #14

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.1142. Min Act: -0.0001

Data Index: 1382393 (Python Code)

Max Activating Token Index: 499

Text #15

Max Range: 9.9616. Min Range: -9.9616

Max Act: 7.3852. Min Act: -0.0000

Data Index: 544671 (C4 (Web Text))

Max Activating Token Index: 342

Text #16

Max Range: 9.9616. Min Range: -9.9616

Max Act: 7.3441. Min Act: -0.0000

Data Index: 19790 (C4 (Web Text))

Max Activating Token Index: 713

Text #17

Max Range: 9.9616. Min Range: -9.9616

Max Act: 8.0257. Min Act: -0.0001

Data Index: 1581731 (Python Code)

Max Activating Token Index: 171

Text #18

Max Range: 9.9616. Min Range: -9.9616

Max Act: 7.8382. Min Act: -0.0000

Data Index: 883218 (C4 (Web Text))

Max Activating Token Index: 7

Text #19

Max Range: 9.9616. Min Range: -9.9616

Max Act: 7.7689. Min Act: -0.0000

Data Index: 437901 (C4 (Web Text))

Max Activating Token Index: 16