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

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

Neuron 2466 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3278. Min Act: -0.0000

Data Index: 1561744 (Python Code)

Max Activating Token Index: 206

Text #1

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3089. Min Act: -0.0000

Data Index: 529327 (C4 (Web Text))

Max Activating Token Index: 574

Text #2

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2975. Min Act: -0.0000

Data Index: 1042148 (C4 (Web Text))

Max Activating Token Index: 570

Text #3

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3125. Min Act: -0.0000

Data Index: 257165 (C4 (Web Text))

Max Activating Token Index: 491

Text #4

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2723. Min Act: -0.0000

Data Index: 800412 (C4 (Web Text))

Max Activating Token Index: 115

Text #5

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3053. Min Act: -0.0000

Data Index: 829338 (C4 (Web Text))

Max Activating Token Index: 149

Text #6

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3078. Min Act: -0.0000

Data Index: 282137 (C4 (Web Text))

Max Activating Token Index: 86

Text #7

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.3029. Min Act: -0.0000

Data Index: 907164 (C4 (Web Text))

Max Activating Token Index: 595

Text #8

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2814. Min Act: -0.0000

Data Index: 861307 (C4 (Web Text))

Max Activating Token Index: 284

Text #9

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2788. Min Act: -0.0000

Data Index: 923944 (C4 (Web Text))

Max Activating Token Index: 515

Text #10

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2943. Min Act: -0.0000

Data Index: 1516850 (Python Code)

Max Activating Token Index: 288

Text #11

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2710. Min Act: -0.0000

Data Index: 298470 (C4 (Web Text))

Max Activating Token Index: 173

Text #12

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2815. Min Act: -0.0000

Data Index: 1546660 (Python Code)

Max Activating Token Index: 228

Text #13

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2777. Min Act: -0.0000

Data Index: 1353272 (C4 (Web Text))

Max Activating Token Index: 84

Text #14

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2893. Min Act: -0.0000

Data Index: 630458 (C4 (Web Text))

Max Activating Token Index: 96

Text #15

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2603. Min Act: -0.0000

Data Index: 1003934 (C4 (Web Text))

Max Activating Token Index: 777

Text #16

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2993. Min Act: -0.0000

Data Index: 816978 (C4 (Web Text))

Max Activating Token Index: 545

Text #17

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2459. Min Act: -0.0000

Data Index: 820516 (C4 (Web Text))

Max Activating Token Index: 437

Text #18

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2723. Min Act: -0.0000

Data Index: 94418 (C4 (Web Text))

Max Activating Token Index: 282

Text #19

Max Range: 0.3278. Min Range: -0.3278

Max Act: 0.2683. Min Act: -0.0000

Data Index: 51913 (C4 (Web Text))

Max Activating Token Index: 331