Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4905 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1297. Min Range: -1.1297

Max Act: 1.1297. Min Act: -0.0000

Data Index: 1148199 (C4 (Web Text))

Max Activating Token Index: 265

Text #1

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.9387. Min Act: -0.0000

Data Index: 1566166 (Python Code)

Max Activating Token Index: 675

Text #2

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.8538. Min Act: -0.0000

Data Index: 1330629 (C4 (Web Text))

Max Activating Token Index: 201

Text #3

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.8898. Min Act: -0.0000

Data Index: 815394 (C4 (Web Text))

Max Activating Token Index: 421

Text #4

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.8535. Min Act: -0.0000

Data Index: 489406 (C4 (Web Text))

Max Activating Token Index: 628

Text #5

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.9530. Min Act: -0.0000

Data Index: 956944 (C4 (Web Text))

Max Activating Token Index: 818

Text #6

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7905. Min Act: -0.0000

Data Index: 1408953 (Python Code)

Max Activating Token Index: 144

Text #7

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7846. Min Act: -0.0000

Data Index: 1496178 (Python Code)

Max Activating Token Index: 293

Text #8

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7925. Min Act: -0.0000

Data Index: 678109 (C4 (Web Text))

Max Activating Token Index: 930

Text #9

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7557. Min Act: -0.0000

Data Index: 1423427 (Python Code)

Max Activating Token Index: 871

Text #10

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7319. Min Act: -0.0000

Data Index: 1386998 (Python Code)

Max Activating Token Index: 524

Text #11

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7065. Min Act: -0.0000

Data Index: 385792 (C4 (Web Text))

Max Activating Token Index: 52

Text #12

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6988. Min Act: -0.0000

Data Index: 1605638 (Python Code)

Max Activating Token Index: 1007

Text #13

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6588. Min Act: -0.0000

Data Index: 706883 (C4 (Web Text))

Max Activating Token Index: 921

Text #14

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6731. Min Act: -0.0000

Data Index: 598279 (C4 (Web Text))

Max Activating Token Index: 934

Text #15

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7487. Min Act: -0.0000

Data Index: 1223968 (C4 (Web Text))

Max Activating Token Index: 368

Text #16

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6504. Min Act: -0.0000

Data Index: 594427 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6910. Min Act: -0.0000

Data Index: 1655961 (Python Code)

Max Activating Token Index: 668

Text #18

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.6952. Min Act: -0.0000

Data Index: 730204 (C4 (Web Text))

Max Activating Token Index: 10

Text #19

Max Range: 1.1297. Min Range: -1.1297

Max Act: 0.7284. Min Act: -0.0000

Data Index: 789325 (C4 (Web Text))

Max Activating Token Index: 16