Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 492 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.0557. Min Range: -1.0557

Max Act: 1.0557. Min Act: -0.0002

Data Index: 585919 (C4 (Web Text))

Max Activating Token Index: 31

Text #1

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.9991. Min Act: -0.0002

Data Index: 164227 (C4 (Web Text))

Max Activating Token Index: 504

Text #2

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8457. Min Act: -0.0002

Data Index: 1326810 (C4 (Web Text))

Max Activating Token Index: 145

Text #3

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8694. Min Act: -0.0002

Data Index: 255171 (C4 (Web Text))

Max Activating Token Index: 954

Text #4

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8792. Min Act: -0.0001

Data Index: 639457 (C4 (Web Text))

Max Activating Token Index: 448

Text #5

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8771. Min Act: -0.0002

Data Index: 163929 (C4 (Web Text))

Max Activating Token Index: 278

Text #6

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8481. Min Act: -0.0002

Data Index: 90445 (C4 (Web Text))

Max Activating Token Index: 851

Text #7

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8456. Min Act: -0.0002

Data Index: 607718 (C4 (Web Text))

Max Activating Token Index: 396

Text #8

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8475. Min Act: -0.0001

Data Index: 363 (C4 (Web Text))

Max Activating Token Index: 1002

Text #9

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8164. Min Act: -0.0002

Data Index: 110363 (C4 (Web Text))

Max Activating Token Index: 44

Text #10

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8375. Min Act: -0.0002

Data Index: 352039 (C4 (Web Text))

Max Activating Token Index: 89

Text #11

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8166. Min Act: -0.0002

Data Index: 843915 (C4 (Web Text))

Max Activating Token Index: 353

Text #12

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8078. Min Act: -0.0001

Data Index: 1078393 (C4 (Web Text))

Max Activating Token Index: 237

Text #13

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.6421. Min Act: -0.0002

Data Index: 1323694 (C4 (Web Text))

Max Activating Token Index: 704

Text #14

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8136. Min Act: -0.0001

Data Index: 85415 (C4 (Web Text))

Max Activating Token Index: 411

Text #15

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.7857. Min Act: -0.0002

Data Index: 630067 (C4 (Web Text))

Max Activating Token Index: 290

Text #16

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.7638. Min Act: -0.0002

Data Index: 1243606 (C4 (Web Text))

Max Activating Token Index: 437

Text #17

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.8095. Min Act: -0.0002

Data Index: 818527 (C4 (Web Text))

Max Activating Token Index: 801

Text #18

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.7403. Min Act: -0.0002

Data Index: 477271 (C4 (Web Text))

Max Activating Token Index: 81

Text #19

Max Range: 1.0557. Min Range: -1.0557

Max Act: 0.7869. Min Act: -0.0002

Data Index: 891939 (C4 (Web Text))

Max Activating Token Index: 491