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

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

Neuron 27 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.5486. Min Act: -0.0001

Data Index: 597317 (C4 (Web Text))

Max Activating Token Index: 44

Text #1

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.5808. Min Act: -0.0001

Data Index: 149999 (C4 (Web Text))

Max Activating Token Index: 511

Text #2

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.4838. Min Act: -0.0001

Data Index: 1211312 (C4 (Web Text))

Max Activating Token Index: 168

Text #3

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.3613. Min Act: -0.0001

Data Index: 491284 (C4 (Web Text))

Max Activating Token Index: 127

Text #4

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.3723. Min Act: -0.0001

Data Index: 760899 (C4 (Web Text))

Max Activating Token Index: 173

Text #5

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.4003. Min Act: -0.0001

Data Index: 407487 (C4 (Web Text))

Max Activating Token Index: 722

Text #6

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2564. Min Act: -0.0000

Data Index: 393136 (C4 (Web Text))

Max Activating Token Index: 684

Text #7

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2810. Min Act: -0.0001

Data Index: 1459705 (Python Code)

Max Activating Token Index: 552

Text #8

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2107. Min Act: -0.0001

Data Index: 105180 (C4 (Web Text))

Max Activating Token Index: 210

Text #9

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2454. Min Act: -0.0001

Data Index: 1209887 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1145. Min Act: -0.0001

Data Index: 93005 (C4 (Web Text))

Max Activating Token Index: 1007

Text #11

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2159. Min Act: -0.0001

Data Index: 874797 (C4 (Web Text))

Max Activating Token Index: 88

Text #12

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1765. Min Act: -0.0000

Data Index: 153920 (C4 (Web Text))

Max Activating Token Index: 5

Text #13

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2666. Min Act: -0.0001

Data Index: 1386158 (Python Code)

Max Activating Token Index: 635

Text #14

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1436. Min Act: -0.0001

Data Index: 1414243 (Python Code)

Max Activating Token Index: 575

Text #15

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.2366. Min Act: -0.0001

Data Index: 1607072 (Python Code)

Max Activating Token Index: 340

Text #16

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1750. Min Act: -0.0001

Data Index: 237169 (C4 (Web Text))

Max Activating Token Index: 98

Text #17

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1656. Min Act: -0.0001

Data Index: 1543425 (Python Code)

Max Activating Token Index: 331

Text #18

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1410. Min Act: -0.0001

Data Index: 1160172 (C4 (Web Text))

Max Activating Token Index: 955

Text #19

Max Range: 1.5808. Min Range: -1.5808

Max Act: 1.1868. Min Act: -0.0001

Data Index: 1583486 (Python Code)

Max Activating Token Index: 650