Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4382 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.5778. Min Range: -2.5778

Max Act: 2.5778. Min Act: -0.0000

Data Index: 1485975 (Python Code)

Max Activating Token Index: 590

Text #1

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.9446. Min Act: -0.0000

Data Index: 1433075 (Python Code)

Max Activating Token Index: 58

Text #2

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.6947. Min Act: -0.0000

Data Index: 1312966 (C4 (Web Text))

Max Activating Token Index: 903

Text #3

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.5785. Min Act: -0.0000

Data Index: 1468349 (Python Code)

Max Activating Token Index: 224

Text #4

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.8262. Min Act: -0.0000

Data Index: 1605543 (Python Code)

Max Activating Token Index: 660

Text #5

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.6030. Min Act: -0.0000

Data Index: 190871 (C4 (Web Text))

Max Activating Token Index: 153

Text #6

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.6562. Min Act: -0.0000

Data Index: 324873 (C4 (Web Text))

Max Activating Token Index: 16

Text #7

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.7394. Min Act: -0.0000

Data Index: 896611 (C4 (Web Text))

Max Activating Token Index: 572

Text #8

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.4613. Min Act: -0.0000

Data Index: 1592680 (Python Code)

Max Activating Token Index: 75

Text #9

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3754. Min Act: -0.0000

Data Index: 891212 (C4 (Web Text))

Max Activating Token Index: 408

Text #10

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.4390. Min Act: -0.0000

Data Index: 477685 (C4 (Web Text))

Max Activating Token Index: 888

Text #11

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.4387. Min Act: -0.0000

Data Index: 1442915 (Python Code)

Max Activating Token Index: 788

Text #12

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.4219. Min Act: -0.0000

Data Index: 1007339 (C4 (Web Text))

Max Activating Token Index: 629

Text #13

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3056. Min Act: -0.0000

Data Index: 1534140 (Python Code)

Max Activating Token Index: 1021

Text #14

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3750. Min Act: -0.0000

Data Index: 1559063 (Python Code)

Max Activating Token Index: 91

Text #15

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3658. Min Act: -0.0000

Data Index: 1390449 (Python Code)

Max Activating Token Index: 241

Text #16

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3628. Min Act: -0.0000

Data Index: 380012 (C4 (Web Text))

Max Activating Token Index: 588

Text #17

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.5614. Min Act: -0.0000

Data Index: 1246837 (C4 (Web Text))

Max Activating Token Index: 536

Text #18

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3160. Min Act: -0.0000

Data Index: 494239 (C4 (Web Text))

Max Activating Token Index: 130

Text #19

Max Range: 2.5778. Min Range: -2.5778

Max Act: 1.3793. Min Act: -0.0000

Data Index: 1203299 (C4 (Web Text))

Max Activating Token Index: 639