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

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

Neuron 1988 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.8955. Min Act: -0.0002

Data Index: 1116649 (C4 (Web Text))

Max Activating Token Index: 122

Text #1

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.7228. Min Act: -0.0003

Data Index: 434703 (C4 (Web Text))

Max Activating Token Index: 328

Text #2

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6480. Min Act: -0.0002

Data Index: 145852 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6464. Min Act: -0.0002

Data Index: 384670 (C4 (Web Text))

Max Activating Token Index: 224

Text #4

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6553. Min Act: -0.0002

Data Index: 332513 (C4 (Web Text))

Max Activating Token Index: 656

Text #5

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6495. Min Act: -0.0003

Data Index: 572037 (C4 (Web Text))

Max Activating Token Index: 35

Text #6

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6208. Min Act: -0.0002

Data Index: 317080 (C4 (Web Text))

Max Activating Token Index: 9

Text #7

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6511. Min Act: -0.0002

Data Index: 470996 (C4 (Web Text))

Max Activating Token Index: 246

Text #8

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5831. Min Act: -0.0002

Data Index: 808603 (C4 (Web Text))

Max Activating Token Index: 459

Text #9

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5978. Min Act: -0.0002

Data Index: 1376299 (Python Code)

Max Activating Token Index: 163

Text #10

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6107. Min Act: -0.0002

Data Index: 1355613 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6502. Min Act: -0.0002

Data Index: 925548 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5529. Min Act: -0.0002

Data Index: 1221453 (C4 (Web Text))

Max Activating Token Index: 801

Text #13

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5750. Min Act: -0.0002

Data Index: 758593 (C4 (Web Text))

Max Activating Token Index: 30

Text #14

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5426. Min Act: -0.0002

Data Index: 1142134 (C4 (Web Text))

Max Activating Token Index: 136

Text #15

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5370. Min Act: -0.0002

Data Index: 1200796 (C4 (Web Text))

Max Activating Token Index: 18

Text #16

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.6171. Min Act: -0.0002

Data Index: 411867 (C4 (Web Text))

Max Activating Token Index: 771

Text #17

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5446. Min Act: -0.0002

Data Index: 1620530 (Python Code)

Max Activating Token Index: 838

Text #18

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5678. Min Act: -0.0002

Data Index: 972034 (C4 (Web Text))

Max Activating Token Index: 11

Text #19

Max Range: 1.8955. Min Range: -1.8955

Max Act: 1.5398. Min Act: -0.0002

Data Index: 1184588 (C4 (Web Text))

Max Activating Token Index: 701