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

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

Neuron 495 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.9650. Min Act: -0.0001

Data Index: 1029594 (C4 (Web Text))

Max Activating Token Index: 439

Text #1

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8629. Min Act: -0.0001

Data Index: 92755 (C4 (Web Text))

Max Activating Token Index: 580

Text #2

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8136. Min Act: -0.0002

Data Index: 185215 (C4 (Web Text))

Max Activating Token Index: 489

Text #3

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8652. Min Act: -0.0002

Data Index: 512977 (C4 (Web Text))

Max Activating Token Index: 550

Text #4

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8632. Min Act: -0.0001

Data Index: 357939 (C4 (Web Text))

Max Activating Token Index: 818

Text #5

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7970. Min Act: -0.0001

Data Index: 60323 (C4 (Web Text))

Max Activating Token Index: 969

Text #6

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8090. Min Act: -0.0002

Data Index: 350568 (C4 (Web Text))

Max Activating Token Index: 960

Text #7

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7458. Min Act: -0.0001

Data Index: 996392 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7850. Min Act: -0.0002

Data Index: 988655 (C4 (Web Text))

Max Activating Token Index: 689

Text #9

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7652. Min Act: -0.0001

Data Index: 661476 (C4 (Web Text))

Max Activating Token Index: 491

Text #10

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8183. Min Act: -0.0001

Data Index: 953306 (C4 (Web Text))

Max Activating Token Index: 417

Text #11

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7966. Min Act: -0.0001

Data Index: 393194 (C4 (Web Text))

Max Activating Token Index: 362

Text #12

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8194. Min Act: -0.0001

Data Index: 339746 (C4 (Web Text))

Max Activating Token Index: 680

Text #13

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7855. Min Act: -0.0002

Data Index: 931078 (C4 (Web Text))

Max Activating Token Index: 349

Text #14

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.8023. Min Act: -0.0002

Data Index: 1069768 (C4 (Web Text))

Max Activating Token Index: 494

Text #15

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7757. Min Act: -0.0001

Data Index: 547594 (C4 (Web Text))

Max Activating Token Index: 50

Text #16

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7432. Min Act: -0.0002

Data Index: 659353 (C4 (Web Text))

Max Activating Token Index: 425

Text #17

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7693. Min Act: -0.0001

Data Index: 512110 (C4 (Web Text))

Max Activating Token Index: 843

Text #18

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7799. Min Act: -0.0001

Data Index: 690700 (C4 (Web Text))

Max Activating Token Index: 386

Text #19

Max Range: 0.9650. Min Range: -0.9650

Max Act: 0.7115. Min Act: -0.0001

Data Index: 759122 (C4 (Web Text))

Max Activating Token Index: 727