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

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

Neuron 4844 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.2547. Min Range: -2.2547

Max Act: 2.2547. Min Act: -0.0000

Data Index: 1356307 (C4 (Web Text))

Max Activating Token Index: 338

Text #1

Max Range: 2.2547. Min Range: -2.2547

Max Act: 2.1339. Min Act: -0.0000

Data Index: 568197 (C4 (Web Text))

Max Activating Token Index: 987

Text #2

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.8206. Min Act: -0.0000

Data Index: 307862 (C4 (Web Text))

Max Activating Token Index: 437

Text #3

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.6291. Min Act: -0.0000

Data Index: 1099641 (C4 (Web Text))

Max Activating Token Index: 10

Text #4

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.7267. Min Act: -0.0000

Data Index: 758186 (C4 (Web Text))

Max Activating Token Index: 141

Text #5

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.5526. Min Act: -0.0000

Data Index: 1187760 (C4 (Web Text))

Max Activating Token Index: 287

Text #6

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.4996. Min Act: -0.0000

Data Index: 283052 (C4 (Web Text))

Max Activating Token Index: 89

Text #7

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.4117. Min Act: -0.0000

Data Index: 636677 (C4 (Web Text))

Max Activating Token Index: 967

Text #8

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.2934. Min Act: -0.0000

Data Index: 1413224 (Python Code)

Max Activating Token Index: 32

Text #9

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1131. Min Act: -0.0000

Data Index: 1656291 (Python Code)

Max Activating Token Index: 423

Text #10

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1541. Min Act: -0.0000

Data Index: 1617098 (Python Code)

Max Activating Token Index: 244

Text #11

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.3597. Min Act: -0.0000

Data Index: 1228805 (C4 (Web Text))

Max Activating Token Index: 494

Text #12

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.2543. Min Act: -0.0000

Data Index: 40814 (C4 (Web Text))

Max Activating Token Index: 556

Text #13

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1394. Min Act: -0.0000

Data Index: 1481275 (Python Code)

Max Activating Token Index: 236

Text #14

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1760. Min Act: -0.0000

Data Index: 850804 (C4 (Web Text))

Max Activating Token Index: 328

Text #15

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.2248. Min Act: -0.0000

Data Index: 786786 (C4 (Web Text))

Max Activating Token Index: 591

Text #16

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1029. Min Act: -0.0000

Data Index: 987111 (C4 (Web Text))

Max Activating Token Index: 167

Text #17

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.2070. Min Act: -0.0000

Data Index: 1304676 (C4 (Web Text))

Max Activating Token Index: 804

Text #18

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1219. Min Act: -0.0000

Data Index: 1632841 (Python Code)

Max Activating Token Index: 220

Text #19

Max Range: 2.2547. Min Range: -2.2547

Max Act: 1.1922. Min Act: -0.0000

Data Index: 1555998 (Python Code)

Max Activating Token Index: 104