Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 256 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0585. Min Act: -0.0003

Data Index: 670093 (C4 (Web Text))

Max Activating Token Index: 577

Text #1

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0569. Min Act: -0.0004

Data Index: 450213 (C4 (Web Text))

Max Activating Token Index: 205

Text #2

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0542. Min Act: -0.0002

Data Index: 989798 (C4 (Web Text))

Max Activating Token Index: 843

Text #3

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0550. Min Act: -0.0003

Data Index: 593347 (C4 (Web Text))

Max Activating Token Index: 710

Text #4

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0535. Min Act: -0.0003

Data Index: 511001 (C4 (Web Text))

Max Activating Token Index: 793

Text #5

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0536. Min Act: -0.0003

Data Index: 722398 (C4 (Web Text))

Max Activating Token Index: 411

Text #6

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0532. Min Act: -0.0003

Data Index: 510875 (C4 (Web Text))

Max Activating Token Index: 592

Text #7

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0511. Min Act: -0.0004

Data Index: 754361 (C4 (Web Text))

Max Activating Token Index: 137

Text #8

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0503. Min Act: -0.0003

Data Index: 104189 (C4 (Web Text))

Max Activating Token Index: 607

Text #9

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0501. Min Act: -0.0003

Data Index: 654653 (C4 (Web Text))

Max Activating Token Index: 363

Text #10

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0503. Min Act: -0.0003

Data Index: 467119 (C4 (Web Text))

Max Activating Token Index: 386

Text #11

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0514. Min Act: -0.0004

Data Index: 535 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0491. Min Act: -0.0003

Data Index: 838256 (C4 (Web Text))

Max Activating Token Index: 883

Text #13

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0483. Min Act: -0.0003

Data Index: 815428 (C4 (Web Text))

Max Activating Token Index: 428

Text #14

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0487. Min Act: -0.0002

Data Index: 220597 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0466. Min Act: -0.0002

Data Index: 875633 (C4 (Web Text))

Max Activating Token Index: 654

Text #16

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0466. Min Act: -0.0003

Data Index: 1191778 (C4 (Web Text))

Max Activating Token Index: 581

Text #17

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0440. Min Act: -0.0004

Data Index: 1173190 (C4 (Web Text))

Max Activating Token Index: 397

Text #18

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0452. Min Act: -0.0002

Data Index: 1302172 (C4 (Web Text))

Max Activating Token Index: 722

Text #19

Max Range: 0.0585. Min Range: -0.0585

Max Act: 0.0463. Min Act: -0.0002

Data Index: 727687 (C4 (Web Text))

Max Activating Token Index: 412