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

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

Neuron 1350 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.2062. Min Act: -0.0001

Data Index: 1194010 (C4 (Web Text))

Max Activating Token Index: 732

Text #1

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1734. Min Act: -0.0001

Data Index: 1315094 (C4 (Web Text))

Max Activating Token Index: 134

Text #2

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1421. Min Act: -0.0001

Data Index: 1044665 (C4 (Web Text))

Max Activating Token Index: 80

Text #3

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0938. Min Act: -0.0001

Data Index: 1252090 (C4 (Web Text))

Max Activating Token Index: 208

Text #4

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1005. Min Act: -0.0001

Data Index: 92570 (C4 (Web Text))

Max Activating Token Index: 41

Text #5

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1235. Min Act: -0.0001

Data Index: 1011687 (C4 (Web Text))

Max Activating Token Index: 269

Text #6

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1032. Min Act: -0.0000

Data Index: 1001387 (C4 (Web Text))

Max Activating Token Index: 967

Text #7

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0910. Min Act: -0.0001

Data Index: 1543984 (Python Code)

Max Activating Token Index: 56

Text #8

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0957. Min Act: -0.0001

Data Index: 1405923 (Python Code)

Max Activating Token Index: 52

Text #9

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1038. Min Act: -0.0001

Data Index: 1586095 (Python Code)

Max Activating Token Index: 765

Text #10

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.1055. Min Act: -0.0001

Data Index: 884534 (C4 (Web Text))

Max Activating Token Index: 988

Text #11

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0962. Min Act: -0.0000

Data Index: 684590 (C4 (Web Text))

Max Activating Token Index: 301

Text #12

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0930. Min Act: -0.0001

Data Index: 604440 (C4 (Web Text))

Max Activating Token Index: 478

Text #13

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0933. Min Act: -0.0001

Data Index: 1393571 (Python Code)

Max Activating Token Index: 40

Text #14

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0921. Min Act: -0.0001

Data Index: 1640514 (Python Code)

Max Activating Token Index: 50

Text #15

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0911. Min Act: -0.0000

Data Index: 101238 (C4 (Web Text))

Max Activating Token Index: 956

Text #16

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0932. Min Act: -0.0000

Data Index: 582128 (C4 (Web Text))

Max Activating Token Index: 187

Text #17

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0857. Min Act: -0.0000

Data Index: 1105254 (C4 (Web Text))

Max Activating Token Index: 853

Text #18

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0831. Min Act: -0.0000

Data Index: 626983 (C4 (Web Text))

Max Activating Token Index: 846

Text #19

Max Range: 0.2062. Min Range: -0.2062

Max Act: 0.0799. Min Act: -0.0001

Data Index: 1124371 (C4 (Web Text))

Max Activating Token Index: 726