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

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

Neuron 4812 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0067. Min Act: -0.0000

Data Index: 484129 (C4 (Web Text))

Max Activating Token Index: 829

Text #1

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0067. Min Act: -0.0000

Data Index: 1307864 (C4 (Web Text))

Max Activating Token Index: 522

Text #2

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0059. Min Act: -0.0000

Data Index: 157705 (C4 (Web Text))

Max Activating Token Index: 7

Text #3

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0059. Min Act: -0.0000

Data Index: 708519 (C4 (Web Text))

Max Activating Token Index: 7

Text #4

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0060. Min Act: -0.0000

Data Index: 1086539 (C4 (Web Text))

Max Activating Token Index: 883

Text #5

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0053. Min Act: -0.0000

Data Index: 1333450 (C4 (Web Text))

Max Activating Token Index: 298

Text #6

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0055. Min Act: -0.0000

Data Index: 14043 (C4 (Web Text))

Max Activating Token Index: 872

Text #7

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0052. Min Act: -0.0000

Data Index: 88482 (C4 (Web Text))

Max Activating Token Index: 85

Text #8

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0048. Min Act: -0.0000

Data Index: 592081 (C4 (Web Text))

Max Activating Token Index: 638

Text #9

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0044. Min Act: -0.0000

Data Index: 262364 (C4 (Web Text))

Max Activating Token Index: 802

Text #10

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0047. Min Act: -0.0000

Data Index: 454236 (C4 (Web Text))

Max Activating Token Index: 13

Text #11

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0048. Min Act: -0.0000

Data Index: 415239 (C4 (Web Text))

Max Activating Token Index: 988

Text #12

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0038. Min Act: -0.0000

Data Index: 882950 (C4 (Web Text))

Max Activating Token Index: 262

Text #13

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0046. Min Act: -0.0000

Data Index: 1220346 (C4 (Web Text))

Max Activating Token Index: 348

Text #14

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0045. Min Act: -0.0000

Data Index: 274331 (C4 (Web Text))

Max Activating Token Index: 6

Text #15

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0041. Min Act: -0.0000

Data Index: 914824 (C4 (Web Text))

Max Activating Token Index: 627

Text #16

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0040. Min Act: -0.0000

Data Index: 1632405 (Python Code)

Max Activating Token Index: 719

Text #17

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0037. Min Act: -0.0000

Data Index: 504890 (C4 (Web Text))

Max Activating Token Index: 158

Text #18

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0042. Min Act: -0.0000

Data Index: 597578 (C4 (Web Text))

Max Activating Token Index: 995

Text #19

Max Range: 0.0067. Min Range: -0.0067

Max Act: 0.0039. Min Act: -0.0000

Data Index: 106256 (C4 (Web Text))

Max Activating Token Index: 935