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

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

Neuron 4269 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3325. Min Act: -0.0000

Data Index: 1482573 (Python Code)

Max Activating Token Index: 698

Text #1

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4544. Min Act: -0.0000

Data Index: 950447 (C4 (Web Text))

Max Activating Token Index: 484

Text #2

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4681. Min Act: -0.0000

Data Index: 309429 (C4 (Web Text))

Max Activating Token Index: 441

Text #3

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4404. Min Act: -0.0000

Data Index: 1535479 (Python Code)

Max Activating Token Index: 311

Text #4

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4199. Min Act: -0.0000

Data Index: 650955 (C4 (Web Text))

Max Activating Token Index: 239

Text #5

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3794. Min Act: -0.0000

Data Index: 1623164 (Python Code)

Max Activating Token Index: 262

Text #6

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4814. Min Act: -0.0000

Data Index: 1592958 (Python Code)

Max Activating Token Index: 317

Text #7

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.1475. Min Act: -0.0000

Data Index: 1593198 (Python Code)

Max Activating Token Index: 298

Text #8

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3532. Min Act: -0.0000

Data Index: 1266659 (C4 (Web Text))

Max Activating Token Index: 297

Text #9

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.2102. Min Act: -0.0000

Data Index: 1498230 (Python Code)

Max Activating Token Index: 305

Text #10

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3140. Min Act: -0.0000

Data Index: 669758 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.4491. Min Act: -0.0000

Data Index: 891836 (C4 (Web Text))

Max Activating Token Index: 399

Text #12

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.1821. Min Act: -0.0000

Data Index: 1414958 (Python Code)

Max Activating Token Index: 625

Text #13

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.2930. Min Act: -0.0000

Data Index: 53007 (C4 (Web Text))

Max Activating Token Index: 188

Text #14

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.1446. Min Act: -0.0000

Data Index: 1620908 (Python Code)

Max Activating Token Index: 147

Text #15

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.2599. Min Act: -0.0000

Data Index: 655196 (C4 (Web Text))

Max Activating Token Index: 56

Text #16

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3975. Min Act: -0.0000

Data Index: 1438702 (Python Code)

Max Activating Token Index: 30

Text #17

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.3621. Min Act: -0.0000

Data Index: 1546905 (Python Code)

Max Activating Token Index: 688

Text #18

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.1205. Min Act: -0.0000

Data Index: 1167040 (C4 (Web Text))

Max Activating Token Index: 656

Text #19

Max Range: 2.4814. Min Range: -2.4814

Max Act: 2.1197. Min Act: -0.0000

Data Index: 1383206 (Python Code)

Max Activating Token Index: 410