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

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

Neuron 2580 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0628. Min Act: -0.0000

Data Index: 629511 (C4 (Web Text))

Max Activating Token Index: 264

Text #1

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0641. Min Act: -0.0000

Data Index: 236294 (C4 (Web Text))

Max Activating Token Index: 514

Text #2

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0605. Min Act: -0.0000

Data Index: 1203987 (C4 (Web Text))

Max Activating Token Index: 258

Text #3

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0530. Min Act: -0.0000

Data Index: 161825 (C4 (Web Text))

Max Activating Token Index: 614

Text #4

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0509. Min Act: -0.0000

Data Index: 1609270 (Python Code)

Max Activating Token Index: 726

Text #5

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0530. Min Act: -0.0000

Data Index: 137946 (C4 (Web Text))

Max Activating Token Index: 146

Text #6

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0530. Min Act: -0.0000

Data Index: 896707 (C4 (Web Text))

Max Activating Token Index: 268

Text #7

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0529. Min Act: -0.0000

Data Index: 1267238 (C4 (Web Text))

Max Activating Token Index: 782

Text #8

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0508. Min Act: -0.0000

Data Index: 989592 (C4 (Web Text))

Max Activating Token Index: 67

Text #9

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0521. Min Act: -0.0000

Data Index: 848568 (C4 (Web Text))

Max Activating Token Index: 214

Text #10

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0519. Min Act: -0.0000

Data Index: 993212 (C4 (Web Text))

Max Activating Token Index: 663

Text #11

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0537. Min Act: -0.0000

Data Index: 396741 (C4 (Web Text))

Max Activating Token Index: 958

Text #12

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0487. Min Act: -0.0000

Data Index: 3697 (C4 (Web Text))

Max Activating Token Index: 525

Text #13

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0484. Min Act: -0.0000

Data Index: 1553316 (Python Code)

Max Activating Token Index: 760

Text #14

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0498. Min Act: -0.0000

Data Index: 610981 (C4 (Web Text))

Max Activating Token Index: 25

Text #15

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0515. Min Act: -0.0000

Data Index: 1548646 (Python Code)

Max Activating Token Index: 918

Text #16

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0494. Min Act: -0.0000

Data Index: 170767 (C4 (Web Text))

Max Activating Token Index: 915

Text #17

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0489. Min Act: -0.0000

Data Index: 1550801 (Python Code)

Max Activating Token Index: 770

Text #18

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0472. Min Act: -0.0000

Data Index: 420160 (C4 (Web Text))

Max Activating Token Index: 1015

Text #19

Max Range: 0.0641. Min Range: -0.0641

Max Act: 0.0484. Min Act: -0.0000

Data Index: 1245943 (C4 (Web Text))

Max Activating Token Index: 1003