Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1224 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0536. Min Act: -0.0000

Data Index: 291743 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0217. Min Act: -0.0001

Data Index: 1215785 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0163. Min Act: -0.0000

Data Index: 1289313 (C4 (Web Text))

Max Activating Token Index: 20

Text #3

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0152. Min Act: -0.0000

Data Index: 939558 (C4 (Web Text))

Max Activating Token Index: 13

Text #4

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0155. Min Act: -0.0000

Data Index: 1468880 (Python Code)

Max Activating Token Index: 8

Text #5

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0142. Min Act: -0.0001

Data Index: 1081945 (C4 (Web Text))

Max Activating Token Index: 27

Text #6

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0149. Min Act: -0.0001

Data Index: 376774 (C4 (Web Text))

Max Activating Token Index: 10

Text #7

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0150. Min Act: -0.0000

Data Index: 70816 (C4 (Web Text))

Max Activating Token Index: 16

Text #8

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0148. Min Act: -0.0000

Data Index: 393312 (C4 (Web Text))

Max Activating Token Index: 14

Text #9

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0143. Min Act: -0.0001

Data Index: 813401 (C4 (Web Text))

Max Activating Token Index: 23

Text #10

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0122. Min Act: -0.0000

Data Index: 1159947 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0139. Min Act: 0.0000

Data Index: 383133 (C4 (Web Text))

Max Activating Token Index: 36

Text #12

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0137. Min Act: -0.0000

Data Index: 939534 (C4 (Web Text))

Max Activating Token Index: 50

Text #13

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0137. Min Act: -0.0000

Data Index: 1379709 (Python Code)

Max Activating Token Index: 39

Text #14

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0136. Min Act: -0.0001

Data Index: 1510468 (Python Code)

Max Activating Token Index: 88

Text #15

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0140. Min Act: -0.0001

Data Index: 769619 (C4 (Web Text))

Max Activating Token Index: 9

Text #16

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0132. Min Act: -0.0000

Data Index: 1605615 (Python Code)

Max Activating Token Index: 28

Text #17

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0137. Min Act: -0.0000

Data Index: 1076997 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0128. Min Act: -0.0000

Data Index: 823410 (C4 (Web Text))

Max Activating Token Index: 24

Text #19

Max Range: 0.0536. Min Range: -0.0536

Max Act: 0.0131. Min Act: -0.0001

Data Index: 621205 (C4 (Web Text))

Max Activating Token Index: 94