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

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

Neuron 4662 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3716. Min Range: -3.3716

Max Act: 3.3716. Min Act: -0.0000

Data Index: 279270 (C4 (Web Text))

Max Activating Token Index: 187

Text #1

Max Range: 3.3716. Min Range: -3.3716

Max Act: 3.0191. Min Act: -0.0000

Data Index: 344991 (C4 (Web Text))

Max Activating Token Index: 152

Text #2

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.7860. Min Act: -0.0000

Data Index: 782933 (C4 (Web Text))

Max Activating Token Index: 222

Text #3

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.7752. Min Act: -0.0000

Data Index: 799318 (C4 (Web Text))

Max Activating Token Index: 662

Text #4

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.6096. Min Act: -0.0000

Data Index: 277268 (C4 (Web Text))

Max Activating Token Index: 233

Text #5

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.7821. Min Act: -0.0000

Data Index: 346656 (C4 (Web Text))

Max Activating Token Index: 712

Text #6

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4898. Min Act: -0.0000

Data Index: 370085 (C4 (Web Text))

Max Activating Token Index: 970

Text #7

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.6099. Min Act: -0.0000

Data Index: 1579847 (Python Code)

Max Activating Token Index: 416

Text #8

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4661. Min Act: -0.0000

Data Index: 265753 (C4 (Web Text))

Max Activating Token Index: 410

Text #9

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.5828. Min Act: -0.0000

Data Index: 384382 (C4 (Web Text))

Max Activating Token Index: 1019

Text #10

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.5621. Min Act: -0.0000

Data Index: 1543640 (Python Code)

Max Activating Token Index: 600

Text #11

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.5312. Min Act: -0.0000

Data Index: 1174253 (C4 (Web Text))

Max Activating Token Index: 430

Text #12

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.5221. Min Act: -0.0000

Data Index: 458015 (C4 (Web Text))

Max Activating Token Index: 456

Text #13

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4970. Min Act: -0.0000

Data Index: 687338 (C4 (Web Text))

Max Activating Token Index: 498

Text #14

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.3691. Min Act: -0.0000

Data Index: 1418015 (Python Code)

Max Activating Token Index: 742

Text #15

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4566. Min Act: -0.0000

Data Index: 1204901 (C4 (Web Text))

Max Activating Token Index: 198

Text #16

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4376. Min Act: -0.0000

Data Index: 283092 (C4 (Web Text))

Max Activating Token Index: 202

Text #17

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4512. Min Act: -0.0000

Data Index: 1609040 (Python Code)

Max Activating Token Index: 596

Text #18

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.3139. Min Act: -0.0000

Data Index: 382070 (C4 (Web Text))

Max Activating Token Index: 386

Text #19

Max Range: 3.3716. Min Range: -3.3716

Max Act: 2.4288. Min Act: -0.0000

Data Index: 1047285 (C4 (Web Text))

Max Activating Token Index: 18