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

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

Neuron 3668 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.7015. Min Act: -0.0000

Data Index: 327122 (C4 (Web Text))

Max Activating Token Index: 719

Text #1

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.5228. Min Act: -0.0000

Data Index: 689322 (C4 (Web Text))

Max Activating Token Index: 869

Text #2

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.5521. Min Act: -0.0000

Data Index: 81273 (C4 (Web Text))

Max Activating Token Index: 993

Text #3

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.5524. Min Act: -0.0000

Data Index: 955627 (C4 (Web Text))

Max Activating Token Index: 510

Text #4

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.3954. Min Act: -0.0000

Data Index: 54430 (C4 (Web Text))

Max Activating Token Index: 734

Text #5

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.5602. Min Act: -0.0000

Data Index: 1272966 (C4 (Web Text))

Max Activating Token Index: 997

Text #6

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.5400. Min Act: -0.0000

Data Index: 1600985 (Python Code)

Max Activating Token Index: 737

Text #7

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.3498. Min Act: -0.0000

Data Index: 236956 (C4 (Web Text))

Max Activating Token Index: 581

Text #8

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.4108. Min Act: -0.0000

Data Index: 232992 (C4 (Web Text))

Max Activating Token Index: 724

Text #9

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.4153. Min Act: -0.0000

Data Index: 1655164 (Python Code)

Max Activating Token Index: 668

Text #10

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.3438. Min Act: -0.0000

Data Index: 1516359 (Python Code)

Max Activating Token Index: 804

Text #11

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.2927. Min Act: -0.0000

Data Index: 389953 (C4 (Web Text))

Max Activating Token Index: 850

Text #12

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.3304. Min Act: -0.0000

Data Index: 1516410 (Python Code)

Max Activating Token Index: 873

Text #13

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.2460. Min Act: -0.0000

Data Index: 28121 (C4 (Web Text))

Max Activating Token Index: 808

Text #14

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.1013. Min Act: -0.0000

Data Index: 1296939 (C4 (Web Text))

Max Activating Token Index: 650

Text #15

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.3902. Min Act: -0.0000

Data Index: 1308124 (C4 (Web Text))

Max Activating Token Index: 916

Text #16

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.2773. Min Act: -0.0000

Data Index: 1389996 (Python Code)

Max Activating Token Index: 615

Text #17

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.2406. Min Act: -0.0000

Data Index: 1353554 (C4 (Web Text))

Max Activating Token Index: 469

Text #18

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.1435. Min Act: -0.0000

Data Index: 1470342 (Python Code)

Max Activating Token Index: 967

Text #19

Max Range: 1.7015. Min Range: -1.7015

Max Act: 1.2660. Min Act: -0.0000

Data Index: 515286 (C4 (Web Text))

Max Activating Token Index: 905