Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4320 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.3081. Min Act: -0.0000

Data Index: 1611378 (Python Code)

Max Activating Token Index: 205

Text #1

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.2173. Min Act: -0.0000

Data Index: 693313 (C4 (Web Text))

Max Activating Token Index: 737

Text #2

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.1844. Min Act: -0.0000

Data Index: 1645467 (Python Code)

Max Activating Token Index: 292

Text #3

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.1620. Min Act: -0.0000

Data Index: 796761 (C4 (Web Text))

Max Activating Token Index: 665

Text #4

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.1347. Min Act: -0.0000

Data Index: 1446266 (Python Code)

Max Activating Token Index: 675

Text #5

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.0993. Min Act: -0.0000

Data Index: 1458895 (Python Code)

Max Activating Token Index: 286

Text #6

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.1012. Min Act: -0.0000

Data Index: 1411223 (Python Code)

Max Activating Token Index: 103

Text #7

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.0804. Min Act: -0.0000

Data Index: 361191 (C4 (Web Text))

Max Activating Token Index: 979

Text #8

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.0398. Min Act: -0.0000

Data Index: 1379776 (Python Code)

Max Activating Token Index: 170

Text #9

Max Range: 2.3081. Min Range: -2.3081

Max Act: 2.1141. Min Act: -0.0000

Data Index: 1213684 (C4 (Web Text))

Max Activating Token Index: 856

Text #10

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9930. Min Act: -0.0000

Data Index: 599536 (C4 (Web Text))

Max Activating Token Index: 699

Text #11

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9926. Min Act: -0.0000

Data Index: 1615949 (Python Code)

Max Activating Token Index: 222

Text #12

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9884. Min Act: -0.0000

Data Index: 1546766 (Python Code)

Max Activating Token Index: 147

Text #13

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9796. Min Act: -0.0000

Data Index: 1475029 (Python Code)

Max Activating Token Index: 59

Text #14

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9698. Min Act: -0.0000

Data Index: 1572720 (Python Code)

Max Activating Token Index: 508

Text #15

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9672. Min Act: -0.0000

Data Index: 1320461 (C4 (Web Text))

Max Activating Token Index: 554

Text #16

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.9321. Min Act: -0.0000

Data Index: 964072 (C4 (Web Text))

Max Activating Token Index: 61

Text #17

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.8966. Min Act: -0.0000

Data Index: 1609487 (Python Code)

Max Activating Token Index: 54

Text #18

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.8888. Min Act: -0.0000

Data Index: 517053 (C4 (Web Text))

Max Activating Token Index: 260

Text #19

Max Range: 2.3081. Min Range: -2.3081

Max Act: 1.7787. Min Act: -0.0000

Data Index: 1460706 (Python Code)

Max Activating Token Index: 907