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

Dataset: The Pile

Neuron 2603 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.5232. Min Range: -9.5232

Max Act: 9.5232. Min Act: -0.0000

Data Index: 90073 (The Pile)

Max Activating Token Index: 607

Text #1

Max Range: 9.5232. Min Range: -9.5232

Max Act: 9.4309. Min Act: -0.0000

Data Index: 365494 (The Pile)

Max Activating Token Index: 429

Text #2

Max Range: 9.5232. Min Range: -9.5232

Max Act: 9.2204. Min Act: -0.0000

Data Index: 946741 (The Pile)

Max Activating Token Index: 655

Text #3

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.8723. Min Act: -0.0000

Data Index: 357092 (The Pile)

Max Activating Token Index: 683

Text #4

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.9395. Min Act: -0.0000

Data Index: 736940 (The Pile)

Max Activating Token Index: 757

Text #5

Max Range: 9.5232. Min Range: -9.5232

Max Act: 9.1680. Min Act: -0.0000

Data Index: 1973636 (The Pile)

Max Activating Token Index: 844

Text #6

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.2354. Min Act: -0.0000

Data Index: 459989 (The Pile)

Max Activating Token Index: 371

Text #7

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.0433. Min Act: -0.0000

Data Index: 1003865 (The Pile)

Max Activating Token Index: 382

Text #8

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.3828. Min Act: -0.0000

Data Index: 803473 (The Pile)

Max Activating Token Index: 380

Text #9

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.4598. Min Act: -0.0000

Data Index: 827161 (The Pile)

Max Activating Token Index: 177

Text #10

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.1538. Min Act: -0.0000

Data Index: 1660432 (The Pile)

Max Activating Token Index: 267

Text #11

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.9123. Min Act: -0.0000

Data Index: 1006059 (The Pile)

Max Activating Token Index: 342

Text #12

Max Range: 9.5232. Min Range: -9.5232

Max Act: 8.0719. Min Act: -0.0000

Data Index: 1196298 (The Pile)

Max Activating Token Index: 341

Text #13

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.1757. Min Act: -0.0000

Data Index: 146964 (The Pile)

Max Activating Token Index: 304

Text #14

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.6199. Min Act: -0.0000

Data Index: 1814634 (The Pile)

Max Activating Token Index: 408

Text #15

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.9541. Min Act: -0.0000

Data Index: 254206 (The Pile)

Max Activating Token Index: 519

Text #16

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.4720. Min Act: -0.0000

Data Index: 1318455 (The Pile)

Max Activating Token Index: 468

Text #17

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.2499. Min Act: -0.0000

Data Index: 1535310 (The Pile)

Max Activating Token Index: 511

Text #18

Max Range: 9.5232. Min Range: -9.5232

Max Act: 7.4483. Min Act: -0.0000

Data Index: 1848468 (The Pile)

Max Activating Token Index: 514

Text #19

Max Range: 9.5232. Min Range: -9.5232

Max Act: 6.8600. Min Act: -0.0000

Data Index: 194098 (The Pile)

Max Activating Token Index: 259