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

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

Neuron 998 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6810. Min Act: -0.0001

Data Index: 614850 (C4 (Web Text))

Max Activating Token Index: 555

Text #1

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6701. Min Act: -0.0001

Data Index: 594559 (C4 (Web Text))

Max Activating Token Index: 85

Text #2

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6467. Min Act: -0.0001

Data Index: 1124394 (C4 (Web Text))

Max Activating Token Index: 48

Text #3

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6465. Min Act: -0.0001

Data Index: 101684 (C4 (Web Text))

Max Activating Token Index: 211

Text #4

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6648. Min Act: -0.0001

Data Index: 214479 (C4 (Web Text))

Max Activating Token Index: 178

Text #5

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6407. Min Act: -0.0001

Data Index: 860180 (C4 (Web Text))

Max Activating Token Index: 84

Text #6

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6687. Min Act: -0.0001

Data Index: 543770 (C4 (Web Text))

Max Activating Token Index: 49

Text #7

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6130. Min Act: -0.0001

Data Index: 1046591 (C4 (Web Text))

Max Activating Token Index: 982

Text #8

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6315. Min Act: -0.0001

Data Index: 576062 (C4 (Web Text))

Max Activating Token Index: 776

Text #9

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6611. Min Act: -0.0001

Data Index: 261582 (C4 (Web Text))

Max Activating Token Index: 75

Text #10

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6307. Min Act: -0.0001

Data Index: 495619 (C4 (Web Text))

Max Activating Token Index: 108

Text #11

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6336. Min Act: -0.0001

Data Index: 1028855 (C4 (Web Text))

Max Activating Token Index: 339

Text #12

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5800. Min Act: -0.0001

Data Index: 355662 (C4 (Web Text))

Max Activating Token Index: 923

Text #13

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5848. Min Act: -0.0001

Data Index: 1055806 (C4 (Web Text))

Max Activating Token Index: 213

Text #14

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5694. Min Act: -0.0001

Data Index: 836840 (C4 (Web Text))

Max Activating Token Index: 680

Text #15

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5982. Min Act: -0.0001

Data Index: 135622 (C4 (Web Text))

Max Activating Token Index: 124

Text #16

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5941. Min Act: -0.0001

Data Index: 1187379 (C4 (Web Text))

Max Activating Token Index: 144

Text #17

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5901. Min Act: -0.0001

Data Index: 420291 (C4 (Web Text))

Max Activating Token Index: 26

Text #18

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.6238. Min Act: -0.0001

Data Index: 1351433 (C4 (Web Text))

Max Activating Token Index: 312

Text #19

Max Range: 0.6810. Min Range: -0.6810

Max Act: 0.5166. Min Act: -0.0001

Data Index: 278236 (C4 (Web Text))

Max Activating Token Index: 529