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

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

Neuron 1458 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.7051. Min Act: -0.0000

Data Index: 1159396 (C4 (Web Text))

Max Activating Token Index: 445

Text #1

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.7735. Min Act: -0.0000

Data Index: 729778 (C4 (Web Text))

Max Activating Token Index: 342

Text #2

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.8168. Min Act: -0.0000

Data Index: 1152676 (C4 (Web Text))

Max Activating Token Index: 940

Text #3

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.7394. Min Act: -0.0000

Data Index: 239020 (C4 (Web Text))

Max Activating Token Index: 524

Text #4

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.7318. Min Act: -0.0000

Data Index: 566776 (C4 (Web Text))

Max Activating Token Index: 610

Text #5

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6770. Min Act: -0.0000

Data Index: 248054 (C4 (Web Text))

Max Activating Token Index: 76

Text #6

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6291. Min Act: -0.0000

Data Index: 428117 (C4 (Web Text))

Max Activating Token Index: 601

Text #7

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6046. Min Act: -0.0000

Data Index: 830265 (C4 (Web Text))

Max Activating Token Index: 822

Text #8

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6500. Min Act: -0.0000

Data Index: 620458 (C4 (Web Text))

Max Activating Token Index: 81

Text #9

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5654. Min Act: -0.0000

Data Index: 52573 (C4 (Web Text))

Max Activating Token Index: 983

Text #10

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5521. Min Act: -0.0000

Data Index: 90866 (C4 (Web Text))

Max Activating Token Index: 871

Text #11

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6602. Min Act: -0.0000

Data Index: 182070 (C4 (Web Text))

Max Activating Token Index: 775

Text #12

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5472. Min Act: -0.0000

Data Index: 1324592 (C4 (Web Text))

Max Activating Token Index: 862

Text #13

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5466. Min Act: -0.0000

Data Index: 1556514 (Python Code)

Max Activating Token Index: 147

Text #14

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.6027. Min Act: -0.0000

Data Index: 1125507 (C4 (Web Text))

Max Activating Token Index: 1011

Text #15

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5884. Min Act: -0.0000

Data Index: 241895 (C4 (Web Text))

Max Activating Token Index: 984

Text #16

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.4995. Min Act: -0.0000

Data Index: 933148 (C4 (Web Text))

Max Activating Token Index: 424

Text #17

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5959. Min Act: -0.0000

Data Index: 1139077 (C4 (Web Text))

Max Activating Token Index: 372

Text #18

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5896. Min Act: -0.0000

Data Index: 395562 (C4 (Web Text))

Max Activating Token Index: 779

Text #19

Max Range: 0.8168. Min Range: -0.8168

Max Act: 0.5262. Min Act: -0.0000

Data Index: 786435 (C4 (Web Text))

Max Activating Token Index: 617