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

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

Neuron 1993 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: 3.5549. Min Range: -3.5549

Max Act: 3.5549. Min Act: -0.0001

Data Index: 1221800 (C4 (Web Text))

Max Activating Token Index: 449

Text #1

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.3872. Min Act: -0.0000

Data Index: 1562318 (Python Code)

Max Activating Token Index: 200

Text #2

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.4772. Min Act: -0.0001

Data Index: 1630442 (Python Code)

Max Activating Token Index: 555

Text #3

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.0430. Min Act: -0.0001

Data Index: 1083493 (C4 (Web Text))

Max Activating Token Index: 596

Text #4

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.2637. Min Act: -0.0001

Data Index: 305293 (C4 (Web Text))

Max Activating Token Index: 151

Text #5

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.1964. Min Act: -0.0001

Data Index: 1354118 (C4 (Web Text))

Max Activating Token Index: 416

Text #6

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.2548. Min Act: -0.0000

Data Index: 1628530 (Python Code)

Max Activating Token Index: 1002

Text #7

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.9353. Min Act: -0.0001

Data Index: 937618 (C4 (Web Text))

Max Activating Token Index: 171

Text #8

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.9139. Min Act: -0.0001

Data Index: 65932 (C4 (Web Text))

Max Activating Token Index: 284

Text #9

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.1514. Min Act: -0.0001

Data Index: 1619931 (Python Code)

Max Activating Token Index: 125

Text #10

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.8824. Min Act: -0.0001

Data Index: 597187 (C4 (Web Text))

Max Activating Token Index: 347

Text #11

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.0171. Min Act: -0.0001

Data Index: 699779 (C4 (Web Text))

Max Activating Token Index: 135

Text #12

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.8409. Min Act: -0.0001

Data Index: 1133586 (C4 (Web Text))

Max Activating Token Index: 143

Text #13

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.9920. Min Act: -0.0000

Data Index: 1554254 (Python Code)

Max Activating Token Index: 977

Text #14

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.1287. Min Act: -0.0001

Data Index: 132109 (C4 (Web Text))

Max Activating Token Index: 42

Text #15

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.6259. Min Act: -0.0001

Data Index: 523588 (C4 (Web Text))

Max Activating Token Index: 406

Text #16

Max Range: 3.5549. Min Range: -3.5549

Max Act: 3.0135. Min Act: -0.0001

Data Index: 383563 (C4 (Web Text))

Max Activating Token Index: 383

Text #17

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.8544. Min Act: -0.0001

Data Index: 343603 (C4 (Web Text))

Max Activating Token Index: 935

Text #18

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.9802. Min Act: -0.0001

Data Index: 804731 (C4 (Web Text))

Max Activating Token Index: 542

Text #19

Max Range: 3.5549. Min Range: -3.5549

Max Act: 2.7085. Min Act: -0.0001

Data Index: 559845 (C4 (Web Text))

Max Activating Token Index: 710