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

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

Neuron 2707 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.7061. Min Act: -0.0001

Data Index: 1555225 (Python Code)

Max Activating Token Index: 663

Text #1

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5444. Min Act: -0.0001

Data Index: 1296300 (C4 (Web Text))

Max Activating Token Index: 547

Text #2

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5540. Min Act: -0.0001

Data Index: 1421755 (Python Code)

Max Activating Token Index: 364

Text #3

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5619. Min Act: -0.0001

Data Index: 1033886 (C4 (Web Text))

Max Activating Token Index: 897

Text #4

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5453. Min Act: -0.0001

Data Index: 1423660 (Python Code)

Max Activating Token Index: 84

Text #5

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5250. Min Act: -0.0001

Data Index: 1639574 (Python Code)

Max Activating Token Index: 15

Text #6

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5372. Min Act: -0.0001

Data Index: 862850 (C4 (Web Text))

Max Activating Token Index: 417

Text #7

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5308. Min Act: -0.0001

Data Index: 1484702 (Python Code)

Max Activating Token Index: 742

Text #8

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5358. Min Act: -0.0001

Data Index: 1522449 (Python Code)

Max Activating Token Index: 794

Text #9

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5176. Min Act: -0.0001

Data Index: 1218724 (C4 (Web Text))

Max Activating Token Index: 27

Text #10

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5311. Min Act: -0.0001

Data Index: 304633 (C4 (Web Text))

Max Activating Token Index: 120

Text #11

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5101. Min Act: -0.0001

Data Index: 1508841 (Python Code)

Max Activating Token Index: 459

Text #12

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5215. Min Act: -0.0001

Data Index: 1420676 (Python Code)

Max Activating Token Index: 21

Text #13

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.4973. Min Act: -0.0001

Data Index: 1436417 (Python Code)

Max Activating Token Index: 636

Text #14

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.4928. Min Act: -0.0001

Data Index: 88770 (C4 (Web Text))

Max Activating Token Index: 458

Text #15

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5072. Min Act: -0.0001

Data Index: 1462235 (Python Code)

Max Activating Token Index: 652

Text #16

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5039. Min Act: -0.0001

Data Index: 179293 (C4 (Web Text))

Max Activating Token Index: 225

Text #17

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5093. Min Act: -0.0001

Data Index: 1502094 (Python Code)

Max Activating Token Index: 653

Text #18

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.4893. Min Act: -0.0001

Data Index: 1368925 (Python Code)

Max Activating Token Index: 281

Text #19

Max Range: 0.7061. Min Range: -0.7061

Max Act: 0.5056. Min Act: -0.0001

Data Index: 1051408 (C4 (Web Text))

Max Activating Token Index: 420