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

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

Neuron 1903 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 4.6144. Min Range: -4.6144

Max Act: 4.6144. Min Act: -0.0000

Data Index: 1586316 (Python Code)

Max Activating Token Index: 413

Text #1

Max Range: 4.6144. Min Range: -4.6144

Max Act: 4.2428. Min Act: -0.0000

Data Index: 1591309 (Python Code)

Max Activating Token Index: 613

Text #2

Max Range: 4.6144. Min Range: -4.6144

Max Act: 4.2220. Min Act: -0.0000

Data Index: 1545161 (Python Code)

Max Activating Token Index: 754

Text #3

Max Range: 4.6144. Min Range: -4.6144

Max Act: 3.8497. Min Act: -0.0000

Data Index: 1562241 (Python Code)

Max Activating Token Index: 917

Text #4

Max Range: 4.6144. Min Range: -4.6144

Max Act: 3.8058. Min Act: -0.0000

Data Index: 1561786 (Python Code)

Max Activating Token Index: 996

Text #5

Max Range: 4.6144. Min Range: -4.6144

Max Act: 3.8992. Min Act: -0.0000

Data Index: 1425548 (Python Code)

Max Activating Token Index: 144

Text #6

Max Range: 4.6144. Min Range: -4.6144

Max Act: 3.0862. Min Act: -0.0000

Data Index: 712853 (C4 (Web Text))

Max Activating Token Index: 432

Text #7

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.9269. Min Act: -0.0000

Data Index: 565911 (C4 (Web Text))

Max Activating Token Index: 251

Text #8

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.7712. Min Act: -0.0000

Data Index: 1644957 (Python Code)

Max Activating Token Index: 971

Text #9

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.8133. Min Act: -0.0000

Data Index: 159281 (C4 (Web Text))

Max Activating Token Index: 421

Text #10

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.4813. Min Act: -0.0000

Data Index: 1034631 (C4 (Web Text))

Max Activating Token Index: 827

Text #11

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.4472. Min Act: -0.0000

Data Index: 1317594 (C4 (Web Text))

Max Activating Token Index: 978

Text #12

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.3006. Min Act: -0.0000

Data Index: 1011900 (C4 (Web Text))

Max Activating Token Index: 79

Text #13

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.2631. Min Act: -0.0000

Data Index: 1284678 (C4 (Web Text))

Max Activating Token Index: 872

Text #14

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.1174. Min Act: -0.0000

Data Index: 653138 (C4 (Web Text))

Max Activating Token Index: 60

Text #15

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.1095. Min Act: -0.0000

Data Index: 518390 (C4 (Web Text))

Max Activating Token Index: 983

Text #16

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.0646. Min Act: -0.0000

Data Index: 1459949 (Python Code)

Max Activating Token Index: 800

Text #17

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.0371. Min Act: -0.0000

Data Index: 970555 (C4 (Web Text))

Max Activating Token Index: 941

Text #18

Max Range: 4.6144. Min Range: -4.6144

Max Act: 1.9122. Min Act: -0.0000

Data Index: 478467 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 4.6144. Min Range: -4.6144

Max Act: 2.0162. Min Act: -0.0000

Data Index: 1036135 (C4 (Web Text))

Max Activating Token Index: 305