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

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

Neuron 1573 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: 1.1070. Min Range: -1.1070

Max Act: 1.1070. Min Act: -0.0001

Data Index: 1330646 (C4 (Web Text))

Max Activating Token Index: 837

Text #1

Max Range: 1.1070. Min Range: -1.1070

Max Act: 1.0651. Min Act: -0.0001

Data Index: 355658 (C4 (Web Text))

Max Activating Token Index: 499

Text #2

Max Range: 1.1070. Min Range: -1.1070

Max Act: 1.0073. Min Act: -0.0001

Data Index: 292603 (C4 (Web Text))

Max Activating Token Index: 328

Text #3

Max Range: 1.1070. Min Range: -1.1070

Max Act: 1.0250. Min Act: -0.0001

Data Index: 723685 (C4 (Web Text))

Max Activating Token Index: 329

Text #4

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9884. Min Act: -0.0001

Data Index: 1240169 (C4 (Web Text))

Max Activating Token Index: 915

Text #5

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9552. Min Act: -0.0001

Data Index: 1571238 (Python Code)

Max Activating Token Index: 895

Text #6

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9631. Min Act: -0.0001

Data Index: 904808 (C4 (Web Text))

Max Activating Token Index: 150

Text #7

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9495. Min Act: -0.0001

Data Index: 1241435 (C4 (Web Text))

Max Activating Token Index: 193

Text #8

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9194. Min Act: -0.0001

Data Index: 456086 (C4 (Web Text))

Max Activating Token Index: 960

Text #9

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9116. Min Act: -0.0001

Data Index: 1275801 (C4 (Web Text))

Max Activating Token Index: 360

Text #10

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9365. Min Act: -0.0001

Data Index: 886624 (C4 (Web Text))

Max Activating Token Index: 876

Text #11

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9631. Min Act: -0.0001

Data Index: 111252 (C4 (Web Text))

Max Activating Token Index: 543

Text #12

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9365. Min Act: -0.0001

Data Index: 1190814 (C4 (Web Text))

Max Activating Token Index: 810

Text #13

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9300. Min Act: -0.0001

Data Index: 1206594 (C4 (Web Text))

Max Activating Token Index: 306

Text #14

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9273. Min Act: -0.0000

Data Index: 1479740 (Python Code)

Max Activating Token Index: 279

Text #15

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9892. Min Act: -0.0001

Data Index: 974511 (C4 (Web Text))

Max Activating Token Index: 788

Text #16

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.8650. Min Act: -0.0001

Data Index: 75245 (C4 (Web Text))

Max Activating Token Index: 737

Text #17

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9460. Min Act: -0.0001

Data Index: 1309193 (C4 (Web Text))

Max Activating Token Index: 30

Text #18

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.8921. Min Act: -0.0001

Data Index: 796116 (C4 (Web Text))

Max Activating Token Index: 987

Text #19

Max Range: 1.1070. Min Range: -1.1070

Max Act: 0.9125. Min Act: -0.0001

Data Index: 446100 (C4 (Web Text))

Max Activating Token Index: 1005