Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3630 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 2.7856. Min Range: -2.7856

Max Act: 2.7856. Min Act: -0.0000

Data Index: 1144642 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 2.7856. Min Range: -2.7856

Max Act: 2.3074. Min Act: -0.0000

Data Index: 168792 (C4 (Web Text))

Max Activating Token Index: 57

Text #2

Max Range: 2.7856. Min Range: -2.7856

Max Act: 2.2503. Min Act: -0.0001

Data Index: 827128 (C4 (Web Text))

Max Activating Token Index: 860

Text #3

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.7871. Min Act: -0.0000

Data Index: 1239331 (C4 (Web Text))

Max Activating Token Index: 756

Text #4

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.8003. Min Act: -0.0000

Data Index: 539111 (C4 (Web Text))

Max Activating Token Index: 580

Text #5

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.5204. Min Act: -0.0000

Data Index: 574899 (C4 (Web Text))

Max Activating Token Index: 30

Text #6

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.6557. Min Act: -0.0000

Data Index: 1313875 (C4 (Web Text))

Max Activating Token Index: 86

Text #7

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.5108. Min Act: -0.0000

Data Index: 1021165 (C4 (Web Text))

Max Activating Token Index: 429

Text #8

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.4400. Min Act: -0.0000

Data Index: 30668 (C4 (Web Text))

Max Activating Token Index: 907

Text #9

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.4086. Min Act: -0.0000

Data Index: 951400 (C4 (Web Text))

Max Activating Token Index: 809

Text #10

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.4567. Min Act: -0.0001

Data Index: 1020306 (C4 (Web Text))

Max Activating Token Index: 751

Text #11

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.3759. Min Act: -0.0001

Data Index: 1273074 (C4 (Web Text))

Max Activating Token Index: 86

Text #12

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.3674. Min Act: -0.0000

Data Index: 1200009 (C4 (Web Text))

Max Activating Token Index: 858

Text #13

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.3826. Min Act: -0.0000

Data Index: 661218 (C4 (Web Text))

Max Activating Token Index: 96

Text #14

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.3544. Min Act: -0.0001

Data Index: 884822 (C4 (Web Text))

Max Activating Token Index: 654

Text #15

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.2972. Min Act: -0.0000

Data Index: 1022741 (C4 (Web Text))

Max Activating Token Index: 616

Text #16

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.3001. Min Act: -0.0000

Data Index: 911009 (C4 (Web Text))

Max Activating Token Index: 214

Text #17

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.2661. Min Act: -0.0000

Data Index: 520068 (C4 (Web Text))

Max Activating Token Index: 299

Text #18

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.2712. Min Act: -0.0001

Data Index: 293743 (C4 (Web Text))

Max Activating Token Index: 343

Text #19

Max Range: 2.7856. Min Range: -2.7856

Max Act: 1.2494. Min Act: -0.0000

Data Index: 1032153 (C4 (Web Text))

Max Activating Token Index: 213