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

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

Neuron 3267 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.7773. Min Act: -0.0000

Data Index: 174734 (C4 (Web Text))

Max Activating Token Index: 283

Text #1

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.8185. Min Act: -0.0000

Data Index: 183786 (C4 (Web Text))

Max Activating Token Index: 638

Text #2

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6991. Min Act: -0.0000

Data Index: 41508 (C4 (Web Text))

Max Activating Token Index: 509

Text #3

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6634. Min Act: -0.0000

Data Index: 1520174 (Python Code)

Max Activating Token Index: 719

Text #4

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.7606. Min Act: -0.0000

Data Index: 957499 (C4 (Web Text))

Max Activating Token Index: 826

Text #5

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6475. Min Act: -0.0000

Data Index: 285240 (C4 (Web Text))

Max Activating Token Index: 877

Text #6

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6905. Min Act: -0.0000

Data Index: 470495 (C4 (Web Text))

Max Activating Token Index: 802

Text #7

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6108. Min Act: -0.0000

Data Index: 1642612 (Python Code)

Max Activating Token Index: 719

Text #8

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6277. Min Act: -0.0000

Data Index: 1464229 (Python Code)

Max Activating Token Index: 431

Text #9

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.6436. Min Act: -0.0000

Data Index: 1317650 (C4 (Web Text))

Max Activating Token Index: 32

Text #10

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5785. Min Act: -0.0000

Data Index: 1350809 (C4 (Web Text))

Max Activating Token Index: 503

Text #11

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5585. Min Act: -0.0000

Data Index: 1535844 (Python Code)

Max Activating Token Index: 635

Text #12

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5957. Min Act: -0.0000

Data Index: 977243 (C4 (Web Text))

Max Activating Token Index: 960

Text #13

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5997. Min Act: -0.0000

Data Index: 1378143 (Python Code)

Max Activating Token Index: 71

Text #14

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5863. Min Act: -0.0000

Data Index: 1139984 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5669. Min Act: -0.0000

Data Index: 1539995 (Python Code)

Max Activating Token Index: 213

Text #16

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5964. Min Act: -0.0000

Data Index: 1606336 (Python Code)

Max Activating Token Index: 224

Text #17

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5457. Min Act: -0.0000

Data Index: 467260 (C4 (Web Text))

Max Activating Token Index: 844

Text #18

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5277. Min Act: -0.0000

Data Index: 1220475 (C4 (Web Text))

Max Activating Token Index: 375

Text #19

Max Range: 0.8185. Min Range: -0.8185

Max Act: 0.5274. Min Act: -0.0000

Data Index: 322719 (C4 (Web Text))

Max Activating Token Index: 32