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

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

Neuron 3964 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2815. Min Act: -0.0000

Data Index: 157646 (C4 (Web Text))

Max Activating Token Index: 317

Text #1

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2965. Min Act: -0.0000

Data Index: 147420 (C4 (Web Text))

Max Activating Token Index: 51

Text #2

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2734. Min Act: -0.0000

Data Index: 980359 (C4 (Web Text))

Max Activating Token Index: 115

Text #3

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2825. Min Act: -0.0001

Data Index: 1028542 (C4 (Web Text))

Max Activating Token Index: 885

Text #4

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2726. Min Act: -0.0000

Data Index: 1043560 (C4 (Web Text))

Max Activating Token Index: 713

Text #5

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2824. Min Act: -0.0000

Data Index: 22283 (C4 (Web Text))

Max Activating Token Index: 218

Text #6

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2622. Min Act: -0.0000

Data Index: 1186899 (C4 (Web Text))

Max Activating Token Index: 911

Text #7

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2723. Min Act: -0.0000

Data Index: 1304439 (C4 (Web Text))

Max Activating Token Index: 912

Text #8

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2536. Min Act: -0.0000

Data Index: 77454 (C4 (Web Text))

Max Activating Token Index: 687

Text #9

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2666. Min Act: -0.0000

Data Index: 1287549 (C4 (Web Text))

Max Activating Token Index: 159

Text #10

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2528. Min Act: -0.0000

Data Index: 407718 (C4 (Web Text))

Max Activating Token Index: 679

Text #11

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2550. Min Act: -0.0001

Data Index: 261549 (C4 (Web Text))

Max Activating Token Index: 36

Text #12

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2456. Min Act: -0.0000

Data Index: 114325 (C4 (Web Text))

Max Activating Token Index: 39

Text #13

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2435. Min Act: -0.0000

Data Index: 658013 (C4 (Web Text))

Max Activating Token Index: 492

Text #14

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2317. Min Act: -0.0000

Data Index: 714269 (C4 (Web Text))

Max Activating Token Index: 921

Text #15

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2630. Min Act: -0.0001

Data Index: 712480 (C4 (Web Text))

Max Activating Token Index: 759

Text #16

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2440. Min Act: -0.0001

Data Index: 375093 (C4 (Web Text))

Max Activating Token Index: 126

Text #17

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2390. Min Act: -0.0000

Data Index: 131605 (C4 (Web Text))

Max Activating Token Index: 746

Text #18

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2462. Min Act: -0.0000

Data Index: 1161599 (C4 (Web Text))

Max Activating Token Index: 518

Text #19

Max Range: 0.2965. Min Range: -0.2965

Max Act: 0.2344. Min Act: -0.0001

Data Index: 1267693 (C4 (Web Text))

Max Activating Token Index: 400