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

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

Neuron 3072 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0692. Min Act: -0.0001

Data Index: 858982 (C4 (Web Text))

Max Activating Token Index: 902

Text #1

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0595. Min Act: -0.0001

Data Index: 1588860 (Python Code)

Max Activating Token Index: 545

Text #2

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0599. Min Act: -0.0001

Data Index: 1448086 (Python Code)

Max Activating Token Index: 124

Text #3

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0607. Min Act: -0.0001

Data Index: 1393466 (Python Code)

Max Activating Token Index: 776

Text #4

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0585. Min Act: -0.0001

Data Index: 1427172 (Python Code)

Max Activating Token Index: 545

Text #5

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0569. Min Act: -0.0001

Data Index: 1285989 (C4 (Web Text))

Max Activating Token Index: 438

Text #6

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0557. Min Act: -0.0001

Data Index: 1042157 (C4 (Web Text))

Max Activating Token Index: 389

Text #7

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0535. Min Act: -0.0001

Data Index: 966387 (C4 (Web Text))

Max Activating Token Index: 267

Text #8

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0561. Min Act: -0.0001

Data Index: 1510582 (Python Code)

Max Activating Token Index: 594

Text #9

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0528. Min Act: -0.0001

Data Index: 1254934 (C4 (Web Text))

Max Activating Token Index: 517

Text #10

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0499. Min Act: -0.0001

Data Index: 492426 (C4 (Web Text))

Max Activating Token Index: 1007

Text #11

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0534. Min Act: -0.0001

Data Index: 373263 (C4 (Web Text))

Max Activating Token Index: 682

Text #12

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0491. Min Act: -0.0001

Data Index: 1026686 (C4 (Web Text))

Max Activating Token Index: 467

Text #13

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0528. Min Act: -0.0001

Data Index: 1482174 (Python Code)

Max Activating Token Index: 527

Text #14

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0476. Min Act: -0.0001

Data Index: 45283 (C4 (Web Text))

Max Activating Token Index: 1013

Text #15

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0500. Min Act: -0.0001

Data Index: 791497 (C4 (Web Text))

Max Activating Token Index: 246

Text #16

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0473. Min Act: -0.0001

Data Index: 1092071 (C4 (Web Text))

Max Activating Token Index: 659

Text #17

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0451. Min Act: -0.0001

Data Index: 316909 (C4 (Web Text))

Max Activating Token Index: 902

Text #18

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0437. Min Act: -0.0001

Data Index: 1230988 (C4 (Web Text))

Max Activating Token Index: 696

Text #19

Max Range: 0.0692. Min Range: -0.0692

Max Act: 0.0468. Min Act: -0.0001

Data Index: 1402177 (Python Code)

Max Activating Token Index: 191