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

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

Neuron 1871 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.5943. Min Act: -0.0001

Data Index: 1619103 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.5984. Min Act: -0.0001

Data Index: 1652796 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.4891. Min Act: -0.0000

Data Index: 1562051 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.4304. Min Act: 0.0000

Data Index: 1438089 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.4215. Min Act: -0.0000

Data Index: 1503303 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.4138. Min Act: -0.0001

Data Index: 1398131 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.2780. Min Act: -0.0001

Data Index: 1234392 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3889. Min Act: -0.0001

Data Index: 1403159 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3513. Min Act: -0.0001

Data Index: 1590009 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3450. Min Act: -0.0001

Data Index: 711099 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3346. Min Act: -0.0001

Data Index: 269510 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3384. Min Act: -0.0001

Data Index: 1547143 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3237. Min Act: -0.0001

Data Index: 1649967 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3145. Min Act: -0.0001

Data Index: 1444595 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3000. Min Act: -0.0001

Data Index: 1614689 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 11.5984. Min Range: -11.5984

Max Act: 10.6549. Min Act: -0.0001

Data Index: 1480623 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.2924. Min Act: -0.0001

Data Index: 1549279 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.0584. Min Act: -0.0000

Data Index: 966628 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.1946. Min Act: -0.0001

Data Index: 1652491 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 11.5984. Min Range: -11.5984

Max Act: 11.3690. Min Act: -0.0000

Data Index: 1642892 (Python Code)

Max Activating Token Index: 1023