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

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

Neuron 4902 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.3948. Min Act: -0.0000

Data Index: 265574 (C4 (Web Text))

Max Activating Token Index: 569

Text #1

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.3440. Min Act: -0.0000

Data Index: 501060 (C4 (Web Text))

Max Activating Token Index: 186

Text #2

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.3135. Min Act: -0.0000

Data Index: 869112 (C4 (Web Text))

Max Activating Token Index: 140

Text #3

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2816. Min Act: -0.0000

Data Index: 999943 (C4 (Web Text))

Max Activating Token Index: 241

Text #4

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2842. Min Act: -0.0000

Data Index: 84360 (C4 (Web Text))

Max Activating Token Index: 609

Text #5

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2667. Min Act: -0.0000

Data Index: 1456172 (Python Code)

Max Activating Token Index: 43

Text #6

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2401. Min Act: -0.0000

Data Index: 209843 (C4 (Web Text))

Max Activating Token Index: 900

Text #7

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2288. Min Act: -0.0000

Data Index: 1119261 (C4 (Web Text))

Max Activating Token Index: 926

Text #8

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2184. Min Act: -0.0000

Data Index: 1297139 (C4 (Web Text))

Max Activating Token Index: 87

Text #9

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2157. Min Act: -0.0000

Data Index: 1524496 (Python Code)

Max Activating Token Index: 855

Text #10

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1956. Min Act: -0.0000

Data Index: 1422667 (Python Code)

Max Activating Token Index: 616

Text #11

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2125. Min Act: -0.0000

Data Index: 1431560 (Python Code)

Max Activating Token Index: 836

Text #12

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1918. Min Act: -0.0000

Data Index: 1631999 (Python Code)

Max Activating Token Index: 89

Text #13

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1957. Min Act: -0.0000

Data Index: 1425221 (Python Code)

Max Activating Token Index: 497

Text #14

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1996. Min Act: -0.0000

Data Index: 1622884 (Python Code)

Max Activating Token Index: 880

Text #15

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1931. Min Act: -0.0000

Data Index: 1316370 (C4 (Web Text))

Max Activating Token Index: 900

Text #16

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.2049. Min Act: 0.0000

Data Index: 236177 (C4 (Web Text))

Max Activating Token Index: 494

Text #17

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1896. Min Act: -0.0000

Data Index: 1561934 (Python Code)

Max Activating Token Index: 853

Text #18

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1958. Min Act: -0.0000

Data Index: 1229773 (C4 (Web Text))

Max Activating Token Index: 222

Text #19

Max Range: 0.3948. Min Range: -0.3948

Max Act: 0.1946. Min Act: -0.0000

Data Index: 1528923 (Python Code)

Max Activating Token Index: 832