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

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

Neuron 4608 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.6230. Min Act: -0.0000

Data Index: 718314 (C4 (Web Text))

Max Activating Token Index: 91

Text #1

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.5637. Min Act: -0.0000

Data Index: 1188614 (C4 (Web Text))

Max Activating Token Index: 358

Text #2

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.5296. Min Act: -0.0000

Data Index: 670655 (C4 (Web Text))

Max Activating Token Index: 767

Text #3

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4678. Min Act: -0.0000

Data Index: 495117 (C4 (Web Text))

Max Activating Token Index: 429

Text #4

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4543. Min Act: -0.0000

Data Index: 1249032 (C4 (Web Text))

Max Activating Token Index: 586

Text #5

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4477. Min Act: -0.0000

Data Index: 894410 (C4 (Web Text))

Max Activating Token Index: 77

Text #6

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4381. Min Act: -0.0000

Data Index: 60009 (C4 (Web Text))

Max Activating Token Index: 465

Text #7

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4471. Min Act: -0.0000

Data Index: 1185013 (C4 (Web Text))

Max Activating Token Index: 733

Text #8

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4312. Min Act: -0.0000

Data Index: 121867 (C4 (Web Text))

Max Activating Token Index: 330

Text #9

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4263. Min Act: -0.0000

Data Index: 577772 (C4 (Web Text))

Max Activating Token Index: 553

Text #10

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4282. Min Act: -0.0000

Data Index: 1067427 (C4 (Web Text))

Max Activating Token Index: 989

Text #11

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4174. Min Act: -0.0000

Data Index: 951089 (C4 (Web Text))

Max Activating Token Index: 822

Text #12

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4114. Min Act: -0.0000

Data Index: 1034975 (C4 (Web Text))

Max Activating Token Index: 697

Text #13

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4105. Min Act: -0.0000

Data Index: 1536768 (Python Code)

Max Activating Token Index: 370

Text #14

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4223. Min Act: -0.0000

Data Index: 787602 (C4 (Web Text))

Max Activating Token Index: 622

Text #15

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4284. Min Act: -0.0000

Data Index: 845607 (C4 (Web Text))

Max Activating Token Index: 794

Text #16

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.3906. Min Act: -0.0000

Data Index: 24448 (C4 (Web Text))

Max Activating Token Index: 615

Text #17

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.3984. Min Act: -0.0000

Data Index: 419399 (C4 (Web Text))

Max Activating Token Index: 69

Text #18

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4110. Min Act: -0.0000

Data Index: 275092 (C4 (Web Text))

Max Activating Token Index: 135

Text #19

Max Range: 0.6230. Min Range: -0.6230

Max Act: 0.4085. Min Act: -0.0000

Data Index: 1172336 (C4 (Web Text))

Max Activating Token Index: 66