Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 857 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2517. Min Act: -0.0001

Data Index: 1475454 (Python Code)

Max Activating Token Index: 152

Text #1

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2341. Min Act: -0.0001

Data Index: 1606905 (Python Code)

Max Activating Token Index: 980

Text #2

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2191. Min Act: -0.0001

Data Index: 515283 (C4 (Web Text))

Max Activating Token Index: 37

Text #3

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2257. Min Act: -0.0001

Data Index: 1301436 (C4 (Web Text))

Max Activating Token Index: 192

Text #4

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2169. Min Act: -0.0001

Data Index: 772821 (C4 (Web Text))

Max Activating Token Index: 250

Text #5

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2156. Min Act: -0.0001

Data Index: 467207 (C4 (Web Text))

Max Activating Token Index: 676

Text #6

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1982. Min Act: -0.0001

Data Index: 512178 (C4 (Web Text))

Max Activating Token Index: 923

Text #7

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2065. Min Act: -0.0001

Data Index: 765410 (C4 (Web Text))

Max Activating Token Index: 143

Text #8

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1959. Min Act: -0.0001

Data Index: 766955 (C4 (Web Text))

Max Activating Token Index: 220

Text #9

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1934. Min Act: -0.0001

Data Index: 1274748 (C4 (Web Text))

Max Activating Token Index: 323

Text #10

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1991. Min Act: -0.0001

Data Index: 822498 (C4 (Web Text))

Max Activating Token Index: 63

Text #11

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.2057. Min Act: -0.0001

Data Index: 428934 (C4 (Web Text))

Max Activating Token Index: 113

Text #12

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1709. Min Act: -0.0001

Data Index: 1386358 (Python Code)

Max Activating Token Index: 588

Text #13

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1986. Min Act: -0.0001

Data Index: 1501093 (Python Code)

Max Activating Token Index: 958

Text #14

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1908. Min Act: -0.0001

Data Index: 367951 (C4 (Web Text))

Max Activating Token Index: 911

Text #15

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1966. Min Act: -0.0001

Data Index: 814149 (C4 (Web Text))

Max Activating Token Index: 33

Text #16

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1908. Min Act: -0.0001

Data Index: 418708 (C4 (Web Text))

Max Activating Token Index: 21

Text #17

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1890. Min Act: -0.0001

Data Index: 209605 (C4 (Web Text))

Max Activating Token Index: 416

Text #18

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1768. Min Act: -0.0001

Data Index: 1174140 (C4 (Web Text))

Max Activating Token Index: 341

Text #19

Max Range: 0.2517. Min Range: -0.2517

Max Act: 0.1803. Min Act: -0.0001

Data Index: 648608 (C4 (Web Text))

Max Activating Token Index: 21