Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 677 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3846. Min Act: -0.0001

Data Index: 1138371 (C4 (Web Text))

Max Activating Token Index: 892

Text #1

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3641. Min Act: -0.0001

Data Index: 631410 (C4 (Web Text))

Max Activating Token Index: 526

Text #2

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3342. Min Act: -0.0001

Data Index: 477986 (C4 (Web Text))

Max Activating Token Index: 207

Text #3

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3608. Min Act: -0.0001

Data Index: 839677 (C4 (Web Text))

Max Activating Token Index: 561

Text #4

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.2782. Min Act: -0.0001

Data Index: 1277911 (C4 (Web Text))

Max Activating Token Index: 898

Text #5

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3214. Min Act: -0.0001

Data Index: 1391563 (Python Code)

Max Activating Token Index: 636

Text #6

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3437. Min Act: -0.0001

Data Index: 292930 (C4 (Web Text))

Max Activating Token Index: 11

Text #7

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3514. Min Act: -0.0001

Data Index: 1005466 (C4 (Web Text))

Max Activating Token Index: 61

Text #8

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3282. Min Act: -0.0001

Data Index: 539033 (C4 (Web Text))

Max Activating Token Index: 697

Text #9

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3588. Min Act: -0.0001

Data Index: 757759 (C4 (Web Text))

Max Activating Token Index: 735

Text #10

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3106. Min Act: -0.0001

Data Index: 1446831 (Python Code)

Max Activating Token Index: 227

Text #11

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.2768. Min Act: -0.0001

Data Index: 405149 (C4 (Web Text))

Max Activating Token Index: 605

Text #12

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3059. Min Act: -0.0001

Data Index: 1046992 (C4 (Web Text))

Max Activating Token Index: 952

Text #13

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3356. Min Act: -0.0001

Data Index: 853737 (C4 (Web Text))

Max Activating Token Index: 309

Text #14

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3181. Min Act: -0.0001

Data Index: 572435 (C4 (Web Text))

Max Activating Token Index: 253

Text #15

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.2796. Min Act: -0.0001

Data Index: 381017 (C4 (Web Text))

Max Activating Token Index: 974

Text #16

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3064. Min Act: -0.0001

Data Index: 289853 (C4 (Web Text))

Max Activating Token Index: 881

Text #17

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3276. Min Act: -0.0001

Data Index: 273032 (C4 (Web Text))

Max Activating Token Index: 461

Text #18

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3316. Min Act: -0.0001

Data Index: 501243 (C4 (Web Text))

Max Activating Token Index: 149

Text #19

Max Range: 0.3846. Min Range: -0.3846

Max Act: 0.3183. Min Act: -0.0001

Data Index: 876874 (C4 (Web Text))

Max Activating Token Index: 12