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

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

Neuron 1800 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.7064. Min Act: -0.0000

Data Index: 392738 (C4 (Web Text))

Max Activating Token Index: 239

Text #1

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.4101. Min Act: -0.0001

Data Index: 90106 (C4 (Web Text))

Max Activating Token Index: 549

Text #2

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.3746. Min Act: -0.0000

Data Index: 1156014 (C4 (Web Text))

Max Activating Token Index: 20

Text #3

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.3526. Min Act: -0.0000

Data Index: 663331 (C4 (Web Text))

Max Activating Token Index: 981

Text #4

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2600. Min Act: -0.0001

Data Index: 1224563 (C4 (Web Text))

Max Activating Token Index: 1017

Text #5

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.3029. Min Act: -0.0000

Data Index: 834986 (C4 (Web Text))

Max Activating Token Index: 546

Text #6

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2463. Min Act: -0.0000

Data Index: 1128450 (C4 (Web Text))

Max Activating Token Index: 52

Text #7

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2081. Min Act: -0.0000

Data Index: 420497 (C4 (Web Text))

Max Activating Token Index: 184

Text #8

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2281. Min Act: -0.0001

Data Index: 1606650 (Python Code)

Max Activating Token Index: 572

Text #9

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2489. Min Act: -0.0000

Data Index: 477834 (C4 (Web Text))

Max Activating Token Index: 698

Text #10

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1779. Min Act: -0.0000

Data Index: 328518 (C4 (Web Text))

Max Activating Token Index: 830

Text #11

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1958. Min Act: -0.0000

Data Index: 880273 (C4 (Web Text))

Max Activating Token Index: 671

Text #12

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2306. Min Act: -0.0001

Data Index: 639931 (C4 (Web Text))

Max Activating Token Index: 918

Text #13

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1558. Min Act: -0.0000

Data Index: 780450 (C4 (Web Text))

Max Activating Token Index: 305

Text #14

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2615. Min Act: -0.0000

Data Index: 1129358 (C4 (Web Text))

Max Activating Token Index: 907

Text #15

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1696. Min Act: -0.0000

Data Index: 867747 (C4 (Web Text))

Max Activating Token Index: 985

Text #16

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1651. Min Act: -0.0001

Data Index: 416229 (C4 (Web Text))

Max Activating Token Index: 913

Text #17

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1997. Min Act: -0.0001

Data Index: 1417842 (Python Code)

Max Activating Token Index: 890

Text #18

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.1919. Min Act: -0.0001

Data Index: 1018689 (C4 (Web Text))

Max Activating Token Index: 896

Text #19

Max Range: 1.7064. Min Range: -1.7064

Max Act: 1.2002. Min Act: -0.0000

Data Index: 1168955 (C4 (Web Text))

Max Activating Token Index: 643