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

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

Neuron 3667 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1490. Min Act: -0.0001

Data Index: 1116384 (C4 (Web Text))

Max Activating Token Index: 174

Text #1

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1343. Min Act: -0.0000

Data Index: 1221481 (C4 (Web Text))

Max Activating Token Index: 429

Text #2

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1260. Min Act: -0.0001

Data Index: 972736 (C4 (Web Text))

Max Activating Token Index: 539

Text #3

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1089. Min Act: -0.0000

Data Index: 646488 (C4 (Web Text))

Max Activating Token Index: 517

Text #4

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1139. Min Act: -0.0001

Data Index: 1188940 (C4 (Web Text))

Max Activating Token Index: 554

Text #5

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1143. Min Act: -0.0001

Data Index: 534375 (C4 (Web Text))

Max Activating Token Index: 456

Text #6

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1141. Min Act: -0.0001

Data Index: 252609 (C4 (Web Text))

Max Activating Token Index: 471

Text #7

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1121. Min Act: -0.0001

Data Index: 372671 (C4 (Web Text))

Max Activating Token Index: 983

Text #8

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1075. Min Act: -0.0001

Data Index: 316552 (C4 (Web Text))

Max Activating Token Index: 881

Text #9

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0979. Min Act: -0.0001

Data Index: 538752 (C4 (Web Text))

Max Activating Token Index: 150

Text #10

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.1001. Min Act: -0.0001

Data Index: 803157 (C4 (Web Text))

Max Activating Token Index: 35

Text #11

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0993. Min Act: -0.0001

Data Index: 1039950 (C4 (Web Text))

Max Activating Token Index: 1000

Text #12

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0975. Min Act: -0.0000

Data Index: 800950 (C4 (Web Text))

Max Activating Token Index: 385

Text #13

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0919. Min Act: -0.0000

Data Index: 908455 (C4 (Web Text))

Max Activating Token Index: 151

Text #14

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0941. Min Act: -0.0000

Data Index: 1070339 (C4 (Web Text))

Max Activating Token Index: 256

Text #15

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0943. Min Act: -0.0000

Data Index: 624311 (C4 (Web Text))

Max Activating Token Index: 465

Text #16

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0973. Min Act: -0.0001

Data Index: 1471296 (Python Code)

Max Activating Token Index: 1008

Text #17

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0982. Min Act: -0.0001

Data Index: 6720 (C4 (Web Text))

Max Activating Token Index: 860

Text #18

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0934. Min Act: -0.0000

Data Index: 796123 (C4 (Web Text))

Max Activating Token Index: 210

Text #19

Max Range: 0.1490. Min Range: -0.1490

Max Act: 0.0903. Min Act: -0.0001

Data Index: 126705 (C4 (Web Text))

Max Activating Token Index: 907