Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3790 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.8464. Min Act: -0.0001

Data Index: 1488277 (Python Code)

Max Activating Token Index: 30

Text #1

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.8440. Min Act: -0.0001

Data Index: 1494058 (Python Code)

Max Activating Token Index: 6

Text #2

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.7583. Min Act: -0.0001

Data Index: 1372637 (Python Code)

Max Activating Token Index: 8

Text #3

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.7583. Min Act: -0.0001

Data Index: 1535133 (Python Code)

Max Activating Token Index: 8

Text #4

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.7583. Min Act: -0.0001

Data Index: 1475471 (Python Code)

Max Activating Token Index: 8

Text #5

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.7382. Min Act: -0.0001

Data Index: 1414730 (Python Code)

Max Activating Token Index: 37

Text #6

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.6081. Min Act: -0.0001

Data Index: 1480597 (Python Code)

Max Activating Token Index: 62

Text #7

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.5162. Min Act: -0.0001

Data Index: 1455787 (Python Code)

Max Activating Token Index: 261

Text #8

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.5772. Min Act: -0.0001

Data Index: 1420864 (Python Code)

Max Activating Token Index: 134

Text #9

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4818. Min Act: -0.0000

Data Index: 1628474 (Python Code)

Max Activating Token Index: 998

Text #10

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4331. Min Act: -0.0001

Data Index: 1373625 (Python Code)

Max Activating Token Index: 836

Text #11

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4710. Min Act: -0.0001

Data Index: 1584377 (Python Code)

Max Activating Token Index: 274

Text #12

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4344. Min Act: -0.0001

Data Index: 661899 (C4 (Web Text))

Max Activating Token Index: 240

Text #13

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.3244. Min Act: -0.0000

Data Index: 1632603 (Python Code)

Max Activating Token Index: 1021

Text #14

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.3631. Min Act: -0.0001

Data Index: 172016 (C4 (Web Text))

Max Activating Token Index: 245

Text #15

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4063. Min Act: -0.0001

Data Index: 1419213 (Python Code)

Max Activating Token Index: 892

Text #16

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.3712. Min Act: -0.0001

Data Index: 1613796 (Python Code)

Max Activating Token Index: 36

Text #17

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4012. Min Act: -0.0001

Data Index: 76686 (C4 (Web Text))

Max Activating Token Index: 716

Text #18

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4088. Min Act: -0.0001

Data Index: 1401551 (Python Code)

Max Activating Token Index: 119

Text #19

Max Range: 1.8464. Min Range: -1.8464

Max Act: 1.4040. Min Act: -0.0001

Data Index: 1578988 (Python Code)

Max Activating Token Index: 36