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

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

Neuron 3626 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.4321. Min Act: -0.0000

Data Index: 620988 (C4 (Web Text))

Max Activating Token Index: 195

Text #1

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.3472. Min Act: -0.0000

Data Index: 489666 (C4 (Web Text))

Max Activating Token Index: 366

Text #2

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.3153. Min Act: -0.0000

Data Index: 1231707 (C4 (Web Text))

Max Activating Token Index: 35

Text #3

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.3064. Min Act: -0.0000

Data Index: 930714 (C4 (Web Text))

Max Activating Token Index: 278

Text #4

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.3068. Min Act: -0.0000

Data Index: 903943 (C4 (Web Text))

Max Activating Token Index: 985

Text #5

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2795. Min Act: -0.0000

Data Index: 200900 (C4 (Web Text))

Max Activating Token Index: 430

Text #6

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2941. Min Act: -0.0000

Data Index: 385315 (C4 (Web Text))

Max Activating Token Index: 470

Text #7

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2697. Min Act: -0.0000

Data Index: 180526 (C4 (Web Text))

Max Activating Token Index: 339

Text #8

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2452. Min Act: -0.0000

Data Index: 1436575 (Python Code)

Max Activating Token Index: 977

Text #9

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2612. Min Act: -0.0000

Data Index: 1196050 (C4 (Web Text))

Max Activating Token Index: 549

Text #10

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2851. Min Act: -0.0000

Data Index: 784626 (C4 (Web Text))

Max Activating Token Index: 897

Text #11

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2487. Min Act: -0.0000

Data Index: 1600331 (Python Code)

Max Activating Token Index: 384

Text #12

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2459. Min Act: -0.0000

Data Index: 198947 (C4 (Web Text))

Max Activating Token Index: 190

Text #13

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2203. Min Act: -0.0000

Data Index: 443454 (C4 (Web Text))

Max Activating Token Index: 249

Text #14

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2166. Min Act: -0.0000

Data Index: 480826 (C4 (Web Text))

Max Activating Token Index: 772

Text #15

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2265. Min Act: -0.0000

Data Index: 1458245 (Python Code)

Max Activating Token Index: 207

Text #16

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2274. Min Act: -0.0000

Data Index: 411997 (C4 (Web Text))

Max Activating Token Index: 965

Text #17

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2100. Min Act: -0.0000

Data Index: 733295 (C4 (Web Text))

Max Activating Token Index: 942

Text #18

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2072. Min Act: -0.0000

Data Index: 1418500 (Python Code)

Max Activating Token Index: 255

Text #19

Max Range: 0.4321. Min Range: -0.4321

Max Act: 0.2186. Min Act: -0.0000

Data Index: 685823 (C4 (Web Text))

Max Activating Token Index: 536