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

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

Neuron 7 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.6305. Min Act: -0.0001

Data Index: 518389 (C4 (Web Text))

Max Activating Token Index: 248

Text #1

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.7341. Min Act: -0.0001

Data Index: 1220200 (C4 (Web Text))

Max Activating Token Index: 52

Text #2

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.6440. Min Act: -0.0001

Data Index: 821476 (C4 (Web Text))

Max Activating Token Index: 844

Text #3

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.5549. Min Act: -0.0001

Data Index: 47051 (C4 (Web Text))

Max Activating Token Index: 113

Text #4

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.4267. Min Act: -0.0001

Data Index: 1453527 (Python Code)

Max Activating Token Index: 236

Text #5

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.4301. Min Act: -0.0001

Data Index: 1562086 (Python Code)

Max Activating Token Index: 266

Text #6

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.4756. Min Act: -0.0001

Data Index: 1202163 (C4 (Web Text))

Max Activating Token Index: 948

Text #7

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.5995. Min Act: -0.0001

Data Index: 1586785 (Python Code)

Max Activating Token Index: 250

Text #8

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.4055. Min Act: -0.0001

Data Index: 306966 (C4 (Web Text))

Max Activating Token Index: 741

Text #9

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3178. Min Act: -0.0001

Data Index: 290134 (C4 (Web Text))

Max Activating Token Index: 859

Text #10

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.2191. Min Act: -0.0001

Data Index: 1631481 (Python Code)

Max Activating Token Index: 92

Text #11

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3706. Min Act: -0.0001

Data Index: 967121 (C4 (Web Text))

Max Activating Token Index: 791

Text #12

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3155. Min Act: -0.0001

Data Index: 370022 (C4 (Web Text))

Max Activating Token Index: 787

Text #13

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.2769. Min Act: -0.0000

Data Index: 1420537 (Python Code)

Max Activating Token Index: 71

Text #14

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3181. Min Act: -0.0001

Data Index: 1075305 (C4 (Web Text))

Max Activating Token Index: 986

Text #15

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3702. Min Act: -0.0001

Data Index: 174452 (C4 (Web Text))

Max Activating Token Index: 964

Text #16

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3057. Min Act: -0.0001

Data Index: 324891 (C4 (Web Text))

Max Activating Token Index: 220

Text #17

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3031. Min Act: -0.0001

Data Index: 591499 (C4 (Web Text))

Max Activating Token Index: 686

Text #18

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.3043. Min Act: -0.0001

Data Index: 965703 (C4 (Web Text))

Max Activating Token Index: 715

Text #19

Max Range: 1.7341. Min Range: -1.7341

Max Act: 1.2352. Min Act: -0.0001

Data Index: 790939 (C4 (Web Text))

Max Activating Token Index: 221