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

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

Neuron 1437 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: 0.6967. Min Range: -0.6967

Max Act: 0.6967. Min Act: -0.0000

Data Index: 1566612 (Python Code)

Max Activating Token Index: 711

Text #1

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6946. Min Act: -0.0000

Data Index: 575176 (C4 (Web Text))

Max Activating Token Index: 434

Text #2

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6730. Min Act: -0.0000

Data Index: 1245355 (C4 (Web Text))

Max Activating Token Index: 37

Text #3

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6884. Min Act: -0.0000

Data Index: 567054 (C4 (Web Text))

Max Activating Token Index: 323

Text #4

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6700. Min Act: -0.0000

Data Index: 1407642 (Python Code)

Max Activating Token Index: 219

Text #5

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6168. Min Act: -0.0000

Data Index: 895950 (C4 (Web Text))

Max Activating Token Index: 927

Text #6

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6150. Min Act: -0.0000

Data Index: 861555 (C4 (Web Text))

Max Activating Token Index: 523

Text #7

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6158. Min Act: -0.0000

Data Index: 1388185 (Python Code)

Max Activating Token Index: 617

Text #8

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5586. Min Act: -0.0000

Data Index: 339001 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5578. Min Act: -0.0000

Data Index: 111437 (C4 (Web Text))

Max Activating Token Index: 965

Text #10

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5638. Min Act: -0.0000

Data Index: 1149620 (C4 (Web Text))

Max Activating Token Index: 152

Text #11

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5551. Min Act: -0.0000

Data Index: 758534 (C4 (Web Text))

Max Activating Token Index: 319

Text #12

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5704. Min Act: -0.0000

Data Index: 957113 (C4 (Web Text))

Max Activating Token Index: 825

Text #13

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5907. Min Act: -0.0000

Data Index: 633270 (C4 (Web Text))

Max Activating Token Index: 736

Text #14

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5549. Min Act: -0.0000

Data Index: 475053 (C4 (Web Text))

Max Activating Token Index: 1018

Text #15

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.6028. Min Act: -0.0000

Data Index: 114517 (C4 (Web Text))

Max Activating Token Index: 479

Text #16

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5641. Min Act: -0.0000

Data Index: 884063 (C4 (Web Text))

Max Activating Token Index: 572

Text #17

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5446. Min Act: -0.0000

Data Index: 619705 (C4 (Web Text))

Max Activating Token Index: 281

Text #18

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5765. Min Act: -0.0000

Data Index: 189337 (C4 (Web Text))

Max Activating Token Index: 488

Text #19

Max Range: 0.6967. Min Range: -0.6967

Max Act: 0.5511. Min Act: -0.0000

Data Index: 1424946 (Python Code)

Max Activating Token Index: 154