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

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

Neuron 4366 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.7740. Min Act: -0.0001

Data Index: 1435843 (Python Code)

Max Activating Token Index: 166

Text #1

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.5636. Min Act: -0.0001

Data Index: 1637666 (Python Code)

Max Activating Token Index: 887

Text #2

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1946. Min Act: -0.0001

Data Index: 1646702 (Python Code)

Max Activating Token Index: 760

Text #3

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1752. Min Act: -0.0001

Data Index: 1371890 (Python Code)

Max Activating Token Index: 49

Text #4

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.3094. Min Act: -0.0001

Data Index: 1420152 (Python Code)

Max Activating Token Index: 137

Text #5

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1492. Min Act: -0.0001

Data Index: 1575486 (Python Code)

Max Activating Token Index: 882

Text #6

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2989. Min Act: -0.0001

Data Index: 1323037 (C4 (Web Text))

Max Activating Token Index: 232

Text #7

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1371. Min Act: -0.0001

Data Index: 1542742 (Python Code)

Max Activating Token Index: 620

Text #8

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1100. Min Act: -0.0001

Data Index: 1420242 (Python Code)

Max Activating Token Index: 299

Text #9

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2506. Min Act: -0.0001

Data Index: 1374979 (Python Code)

Max Activating Token Index: 326

Text #10

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.0979. Min Act: -0.0001

Data Index: 1600361 (Python Code)

Max Activating Token Index: 212

Text #11

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.3873. Min Act: -0.0001

Data Index: 726090 (C4 (Web Text))

Max Activating Token Index: 997

Text #12

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2338. Min Act: -0.0001

Data Index: 1623400 (Python Code)

Max Activating Token Index: 12

Text #13

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.0674. Min Act: -0.0001

Data Index: 1029109 (C4 (Web Text))

Max Activating Token Index: 900

Text #14

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.3350. Min Act: -0.0001

Data Index: 1638503 (Python Code)

Max Activating Token Index: 341

Text #15

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.1867. Min Act: -0.0001

Data Index: 719616 (C4 (Web Text))

Max Activating Token Index: 391

Text #16

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2674. Min Act: -0.0001

Data Index: 1027361 (C4 (Web Text))

Max Activating Token Index: 935

Text #17

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2572. Min Act: -0.0001

Data Index: 1468013 (Python Code)

Max Activating Token Index: 771

Text #18

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.2508. Min Act: -0.0001

Data Index: 1572981 (Python Code)

Max Activating Token Index: 910

Text #19

Max Range: 3.7740. Min Range: -3.7740

Max Act: 3.0923. Min Act: -0.0001

Data Index: 1565753 (Python Code)

Max Activating Token Index: 109