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

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

Neuron 4015 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1531. Min Act: -0.0000

Data Index: 450501 (C4 (Web Text))

Max Activating Token Index: 142

Text #1

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1543. Min Act: -0.0000

Data Index: 1297528 (C4 (Web Text))

Max Activating Token Index: 1016

Text #2

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1490. Min Act: -0.0000

Data Index: 366599 (C4 (Web Text))

Max Activating Token Index: 192

Text #3

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1493. Min Act: -0.0000

Data Index: 705828 (C4 (Web Text))

Max Activating Token Index: 635

Text #4

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1340. Min Act: -0.0000

Data Index: 43196 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1363. Min Act: -0.0000

Data Index: 87545 (C4 (Web Text))

Max Activating Token Index: 36

Text #6

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1372. Min Act: -0.0000

Data Index: 1525495 (Python Code)

Max Activating Token Index: 728

Text #7

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1368. Min Act: -0.0000

Data Index: 1479665 (Python Code)

Max Activating Token Index: 446

Text #8

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1338. Min Act: -0.0000

Data Index: 1116426 (C4 (Web Text))

Max Activating Token Index: 998

Text #9

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1402. Min Act: -0.0000

Data Index: 796738 (C4 (Web Text))

Max Activating Token Index: 88

Text #10

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1259. Min Act: -0.0000

Data Index: 937911 (C4 (Web Text))

Max Activating Token Index: 393

Text #11

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1350. Min Act: -0.0000

Data Index: 1552174 (Python Code)

Max Activating Token Index: 795

Text #12

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1302. Min Act: -0.0000

Data Index: 1291444 (C4 (Web Text))

Max Activating Token Index: 23

Text #13

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1307. Min Act: -0.0001

Data Index: 1432669 (Python Code)

Max Activating Token Index: 414

Text #14

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1324. Min Act: -0.0000

Data Index: 1392255 (Python Code)

Max Activating Token Index: 825

Text #15

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1300. Min Act: -0.0000

Data Index: 716748 (C4 (Web Text))

Max Activating Token Index: 488

Text #16

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1229. Min Act: -0.0000

Data Index: 886329 (C4 (Web Text))

Max Activating Token Index: 381

Text #17

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1232. Min Act: -0.0000

Data Index: 1114231 (C4 (Web Text))

Max Activating Token Index: 127

Text #18

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1144. Min Act: -0.0000

Data Index: 548563 (C4 (Web Text))

Max Activating Token Index: 435

Text #19

Max Range: 0.1543. Min Range: -0.1543

Max Act: 0.1214. Min Act: -0.0000

Data Index: 925137 (C4 (Web Text))

Max Activating Token Index: 552