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

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

Neuron 1710 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: 0.1735. Min Range: -0.1735

Max Act: 0.1735. Min Act: -0.0001

Data Index: 1199010 (C4 (Web Text))

Max Activating Token Index: 540

Text #1

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1437. Min Act: -0.0001

Data Index: 806261 (C4 (Web Text))

Max Activating Token Index: 520

Text #2

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1332. Min Act: -0.0001

Data Index: 6710 (C4 (Web Text))

Max Activating Token Index: 910

Text #3

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1214. Min Act: -0.0001

Data Index: 738134 (C4 (Web Text))

Max Activating Token Index: 960

Text #4

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1327. Min Act: -0.0001

Data Index: 735693 (C4 (Web Text))

Max Activating Token Index: 534

Text #5

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1121. Min Act: -0.0001

Data Index: 1290534 (C4 (Web Text))

Max Activating Token Index: 629

Text #6

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1170. Min Act: -0.0001

Data Index: 634561 (C4 (Web Text))

Max Activating Token Index: 189

Text #7

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1171. Min Act: 0.0002

Data Index: 425650 (C4 (Web Text))

Max Activating Token Index: 251

Text #8

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1147. Min Act: -0.0001

Data Index: 38950 (C4 (Web Text))

Max Activating Token Index: 596

Text #9

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1092. Min Act: -0.0001

Data Index: 951309 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1145. Min Act: -0.0001

Data Index: 16729 (C4 (Web Text))

Max Activating Token Index: 615

Text #11

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1148. Min Act: -0.0000

Data Index: 268506 (C4 (Web Text))

Max Activating Token Index: 575

Text #12

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1132. Min Act: -0.0001

Data Index: 65613 (C4 (Web Text))

Max Activating Token Index: 998

Text #13

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1083. Min Act: -0.0001

Data Index: 1176072 (C4 (Web Text))

Max Activating Token Index: 84

Text #14

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1092. Min Act: -0.0001

Data Index: 50962 (C4 (Web Text))

Max Activating Token Index: 542

Text #15

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1092. Min Act: -0.0000

Data Index: 1367512 (Python Code)

Max Activating Token Index: 891

Text #16

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1039. Min Act: -0.0001

Data Index: 288991 (C4 (Web Text))

Max Activating Token Index: 668

Text #17

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1058. Min Act: -0.0001

Data Index: 582551 (C4 (Web Text))

Max Activating Token Index: 749

Text #18

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1070. Min Act: -0.0001

Data Index: 586614 (C4 (Web Text))

Max Activating Token Index: 849

Text #19

Max Range: 0.1735. Min Range: -0.1735

Max Act: 0.1051. Min Act: -0.0001

Data Index: 1035768 (C4 (Web Text))

Max Activating Token Index: 150