Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1664 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1633. Min Act: -0.0003

Data Index: 269245 (C4 (Web Text))

Max Activating Token Index: 245

Text #1

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1511. Min Act: -0.0003

Data Index: 120467 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1335. Min Act: -0.0002

Data Index: 310534 (C4 (Web Text))

Max Activating Token Index: 32

Text #3

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1364. Min Act: -0.0002

Data Index: 1163343 (C4 (Web Text))

Max Activating Token Index: 579

Text #4

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1343. Min Act: -0.0003

Data Index: 685604 (C4 (Web Text))

Max Activating Token Index: 832

Text #5

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1322. Min Act: -0.0002

Data Index: 441223 (C4 (Web Text))

Max Activating Token Index: 738

Text #6

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1290. Min Act: -0.0002

Data Index: 584198 (C4 (Web Text))

Max Activating Token Index: 63

Text #7

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1270. Min Act: -0.0002

Data Index: 1194416 (C4 (Web Text))

Max Activating Token Index: 920

Text #8

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1237. Min Act: -0.0002

Data Index: 531645 (C4 (Web Text))

Max Activating Token Index: 597

Text #9

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1172. Min Act: -0.0002

Data Index: 80602 (C4 (Web Text))

Max Activating Token Index: 91

Text #10

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1282. Min Act: -0.0002

Data Index: 1103802 (C4 (Web Text))

Max Activating Token Index: 41

Text #11

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1199. Min Act: -0.0002

Data Index: 377208 (C4 (Web Text))

Max Activating Token Index: 50

Text #12

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1191. Min Act: -0.0002

Data Index: 340305 (C4 (Web Text))

Max Activating Token Index: 68

Text #13

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1225. Min Act: -0.0002

Data Index: 1224908 (C4 (Web Text))

Max Activating Token Index: 851

Text #14

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1221. Min Act: -0.0002

Data Index: 631790 (C4 (Web Text))

Max Activating Token Index: 521

Text #15

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1136. Min Act: -0.0002

Data Index: 406022 (C4 (Web Text))

Max Activating Token Index: 64

Text #16

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1126. Min Act: -0.0002

Data Index: 237041 (C4 (Web Text))

Max Activating Token Index: 951

Text #17

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1166. Min Act: -0.0002

Data Index: 668557 (C4 (Web Text))

Max Activating Token Index: 24

Text #18

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1088. Min Act: -0.0002

Data Index: 955437 (C4 (Web Text))

Max Activating Token Index: 529

Text #19

Max Range: 0.1633. Min Range: -0.1633

Max Act: 0.1142. Min Act: -0.0002

Data Index: 1026228 (C4 (Web Text))

Max Activating Token Index: 422