Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4798 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1207. Min Range: -1.1207

Max Act: 1.1207. Min Act: -0.0000

Data Index: 312301 (C4 (Web Text))

Max Activating Token Index: 674

Text #1

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8836. Min Act: -0.0000

Data Index: 607689 (C4 (Web Text))

Max Activating Token Index: 82

Text #2

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8697. Min Act: -0.0000

Data Index: 1479796 (Python Code)

Max Activating Token Index: 197

Text #3

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8354. Min Act: -0.0000

Data Index: 1583716 (Python Code)

Max Activating Token Index: 228

Text #4

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8287. Min Act: -0.0000

Data Index: 1580724 (Python Code)

Max Activating Token Index: 312

Text #5

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8041. Min Act: -0.0000

Data Index: 1593709 (Python Code)

Max Activating Token Index: 729

Text #6

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8027. Min Act: -0.0000

Data Index: 248687 (C4 (Web Text))

Max Activating Token Index: 705

Text #7

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8565. Min Act: -0.0000

Data Index: 384946 (C4 (Web Text))

Max Activating Token Index: 79

Text #8

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7936. Min Act: -0.0000

Data Index: 448564 (C4 (Web Text))

Max Activating Token Index: 99

Text #9

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8200. Min Act: -0.0000

Data Index: 668953 (C4 (Web Text))

Max Activating Token Index: 505

Text #10

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8190. Min Act: -0.0000

Data Index: 300188 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8091. Min Act: -0.0000

Data Index: 12873 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8561. Min Act: -0.0000

Data Index: 333672 (C4 (Web Text))

Max Activating Token Index: 483

Text #13

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.8038. Min Act: -0.0000

Data Index: 898025 (C4 (Web Text))

Max Activating Token Index: 145

Text #14

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7046. Min Act: -0.0000

Data Index: 1468570 (Python Code)

Max Activating Token Index: 875

Text #15

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7902. Min Act: -0.0000

Data Index: 906904 (C4 (Web Text))

Max Activating Token Index: 171

Text #16

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7855. Min Act: -0.0000

Data Index: 1551224 (Python Code)

Max Activating Token Index: 668

Text #17

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7869. Min Act: -0.0000

Data Index: 40796 (C4 (Web Text))

Max Activating Token Index: 185

Text #18

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7780. Min Act: -0.0000

Data Index: 1453042 (Python Code)

Max Activating Token Index: 612

Text #19

Max Range: 1.1207. Min Range: -1.1207

Max Act: 0.7445. Min Act: -0.0000

Data Index: 738589 (C4 (Web Text))

Max Activating Token Index: 58