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

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

Neuron 2231 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3494. Min Act: -0.0001

Data Index: 1462970 (Python Code)

Max Activating Token Index: 645

Text #1

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3340. Min Act: -0.0001

Data Index: 1419292 (Python Code)

Max Activating Token Index: 15

Text #2

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3394. Min Act: -0.0000

Data Index: 148458 (C4 (Web Text))

Max Activating Token Index: 175

Text #3

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3210. Min Act: -0.0001

Data Index: 1284781 (C4 (Web Text))

Max Activating Token Index: 985

Text #4

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3165. Min Act: -0.0000

Data Index: 1561089 (Python Code)

Max Activating Token Index: 6

Text #5

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3215. Min Act: -0.0001

Data Index: 1579088 (Python Code)

Max Activating Token Index: 389

Text #6

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3192. Min Act: -0.0001

Data Index: 1041165 (C4 (Web Text))

Max Activating Token Index: 58

Text #7

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3072. Min Act: -0.0001

Data Index: 1437873 (Python Code)

Max Activating Token Index: 375

Text #8

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3156. Min Act: -0.0000

Data Index: 1996 (C4 (Web Text))

Max Activating Token Index: 260

Text #9

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3043. Min Act: -0.0001

Data Index: 460476 (C4 (Web Text))

Max Activating Token Index: 90

Text #10

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3016. Min Act: -0.0001

Data Index: 419739 (C4 (Web Text))

Max Activating Token Index: 943

Text #11

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3050. Min Act: -0.0001

Data Index: 1624542 (Python Code)

Max Activating Token Index: 109

Text #12

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2938. Min Act: -0.0001

Data Index: 906640 (C4 (Web Text))

Max Activating Token Index: 815

Text #13

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2928. Min Act: -0.0001

Data Index: 443546 (C4 (Web Text))

Max Activating Token Index: 39

Text #14

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2835. Min Act: -0.0001

Data Index: 121293 (C4 (Web Text))

Max Activating Token Index: 166

Text #15

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.3024. Min Act: -0.0001

Data Index: 1517789 (Python Code)

Max Activating Token Index: 528

Text #16

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2900. Min Act: -0.0001

Data Index: 1231752 (C4 (Web Text))

Max Activating Token Index: 110

Text #17

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2739. Min Act: -0.0001

Data Index: 1633050 (Python Code)

Max Activating Token Index: 182

Text #18

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2808. Min Act: -0.0001

Data Index: 677723 (C4 (Web Text))

Max Activating Token Index: 39

Text #19

Max Range: 0.3494. Min Range: -0.3494

Max Act: 0.2662. Min Act: -0.0001

Data Index: 982787 (C4 (Web Text))

Max Activating Token Index: 132