Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 60 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.2368. Min Range: -1.2368

Max Act: 1.2368. Min Act: -0.0001

Data Index: 96383 (C4 (Web Text))

Max Activating Token Index: 1018

Text #1

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.8608. Min Act: -0.0001

Data Index: 412333 (C4 (Web Text))

Max Activating Token Index: 672

Text #2

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.8775. Min Act: -0.0001

Data Index: 715907 (C4 (Web Text))

Max Activating Token Index: 658

Text #3

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.8610. Min Act: -0.0001

Data Index: 248308 (C4 (Web Text))

Max Activating Token Index: 844

Text #4

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.8044. Min Act: -0.0001

Data Index: 421260 (C4 (Web Text))

Max Activating Token Index: 310

Text #5

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7704. Min Act: -0.0001

Data Index: 1515039 (Python Code)

Max Activating Token Index: 92

Text #6

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7469. Min Act: -0.0001

Data Index: 1144999 (C4 (Web Text))

Max Activating Token Index: 120

Text #7

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7669. Min Act: -0.0001

Data Index: 291158 (C4 (Web Text))

Max Activating Token Index: 842

Text #8

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7119. Min Act: -0.0001

Data Index: 1607469 (Python Code)

Max Activating Token Index: 504

Text #9

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7315. Min Act: -0.0001

Data Index: 315815 (C4 (Web Text))

Max Activating Token Index: 672

Text #10

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.7163. Min Act: -0.0001

Data Index: 1104636 (C4 (Web Text))

Max Activating Token Index: 582

Text #11

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6629. Min Act: -0.0001

Data Index: 1179501 (C4 (Web Text))

Max Activating Token Index: 779

Text #12

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6957. Min Act: -0.0001

Data Index: 1612864 (Python Code)

Max Activating Token Index: 628

Text #13

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6869. Min Act: -0.0001

Data Index: 835605 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6620. Min Act: -0.0001

Data Index: 1615256 (Python Code)

Max Activating Token Index: 842

Text #15

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6586. Min Act: -0.0001

Data Index: 1178472 (C4 (Web Text))

Max Activating Token Index: 394

Text #16

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6793. Min Act: -0.0001

Data Index: 1423335 (Python Code)

Max Activating Token Index: 160

Text #17

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6869. Min Act: -0.0001

Data Index: 438969 (C4 (Web Text))

Max Activating Token Index: 328

Text #18

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6670. Min Act: -0.0001

Data Index: 623474 (C4 (Web Text))

Max Activating Token Index: 948

Text #19

Max Range: 1.2368. Min Range: -1.2368

Max Act: 0.6633. Min Act: -0.0001

Data Index: 1623575 (Python Code)

Max Activating Token Index: 72