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

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

Neuron 2372 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.8092. Min Act: -0.0003

Data Index: 713435 (C4 (Web Text))

Max Activating Token Index: 507

Text #1

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.7941. Min Act: -0.0003

Data Index: 436598 (C4 (Web Text))

Max Activating Token Index: 100

Text #2

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.8045. Min Act: -0.0003

Data Index: 545498 (C4 (Web Text))

Max Activating Token Index: 519

Text #3

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6739. Min Act: -0.0003

Data Index: 262485 (C4 (Web Text))

Max Activating Token Index: 296

Text #4

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.7547. Min Act: -0.0003

Data Index: 1240429 (C4 (Web Text))

Max Activating Token Index: 571

Text #5

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.5369. Min Act: -0.0003

Data Index: 1129076 (C4 (Web Text))

Max Activating Token Index: 597

Text #6

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6441. Min Act: -0.0003

Data Index: 898556 (C4 (Web Text))

Max Activating Token Index: 795

Text #7

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6841. Min Act: -0.0003

Data Index: 1615084 (Python Code)

Max Activating Token Index: 655

Text #8

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6757. Min Act: -0.0003

Data Index: 400993 (C4 (Web Text))

Max Activating Token Index: 528

Text #9

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6437. Min Act: -0.0003

Data Index: 1046236 (C4 (Web Text))

Max Activating Token Index: 562

Text #10

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6555. Min Act: -0.0003

Data Index: 1003040 (C4 (Web Text))

Max Activating Token Index: 254

Text #11

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.5908. Min Act: -0.0003

Data Index: 1181884 (C4 (Web Text))

Max Activating Token Index: 549

Text #12

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6777. Min Act: -0.0003

Data Index: 947134 (C4 (Web Text))

Max Activating Token Index: 424

Text #13

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6552. Min Act: -0.0003

Data Index: 47858 (C4 (Web Text))

Max Activating Token Index: 108

Text #14

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6518. Min Act: -0.0003

Data Index: 859745 (C4 (Web Text))

Max Activating Token Index: 1009

Text #15

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6661. Min Act: -0.0003

Data Index: 354738 (C4 (Web Text))

Max Activating Token Index: 33

Text #16

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6660. Min Act: -0.0003

Data Index: 370760 (C4 (Web Text))

Max Activating Token Index: 247

Text #17

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6874. Min Act: -0.0003

Data Index: 321198 (C4 (Web Text))

Max Activating Token Index: 205

Text #18

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.5804. Min Act: -0.0003

Data Index: 445861 (C4 (Web Text))

Max Activating Token Index: 399

Text #19

Max Range: 0.8092. Min Range: -0.8092

Max Act: 0.6844. Min Act: -0.0003

Data Index: 706811 (C4 (Web Text))

Max Activating Token Index: 414