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

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

Neuron 2595 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1485. Min Act: -0.0001

Data Index: 1654857 (Python Code)

Max Activating Token Index: 690

Text #1

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1530. Min Act: -0.0001

Data Index: 270238 (C4 (Web Text))

Max Activating Token Index: 895

Text #2

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1433. Min Act: -0.0001

Data Index: 1425637 (Python Code)

Max Activating Token Index: 674

Text #3

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1232. Min Act: -0.0001

Data Index: 205110 (C4 (Web Text))

Max Activating Token Index: 54

Text #4

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1182. Min Act: -0.0001

Data Index: 171524 (C4 (Web Text))

Max Activating Token Index: 81

Text #5

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1226. Min Act: -0.0001

Data Index: 1403256 (Python Code)

Max Activating Token Index: 482

Text #6

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1132. Min Act: -0.0001

Data Index: 179488 (C4 (Web Text))

Max Activating Token Index: 142

Text #7

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1046. Min Act: -0.0001

Data Index: 632894 (C4 (Web Text))

Max Activating Token Index: 627

Text #8

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1205. Min Act: -0.0001

Data Index: 938559 (C4 (Web Text))

Max Activating Token Index: 750

Text #9

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1114. Min Act: -0.0001

Data Index: 271682 (C4 (Web Text))

Max Activating Token Index: 614

Text #10

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1199. Min Act: -0.0001

Data Index: 1296566 (C4 (Web Text))

Max Activating Token Index: 777

Text #11

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1133. Min Act: -0.0001

Data Index: 795414 (C4 (Web Text))

Max Activating Token Index: 195

Text #12

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1169. Min Act: -0.0001

Data Index: 1462307 (Python Code)

Max Activating Token Index: 978

Text #13

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1014. Min Act: -0.0001

Data Index: 1417213 (Python Code)

Max Activating Token Index: 642

Text #14

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1081. Min Act: -0.0001

Data Index: 537248 (C4 (Web Text))

Max Activating Token Index: 206

Text #15

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1124. Min Act: -0.0001

Data Index: 144975 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1115. Min Act: -0.0001

Data Index: 102165 (C4 (Web Text))

Max Activating Token Index: 581

Text #17

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1162. Min Act: -0.0001

Data Index: 17159 (C4 (Web Text))

Max Activating Token Index: 926

Text #18

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1075. Min Act: -0.0001

Data Index: 391541 (C4 (Web Text))

Max Activating Token Index: 932

Text #19

Max Range: 0.1530. Min Range: -0.1530

Max Act: 0.1110. Min Act: -0.0001

Data Index: 389380 (C4 (Web Text))

Max Activating Token Index: 115