Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 746 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 2.0250. Min Range: -2.0250

Max Act: 2.0250. Min Act: -0.0001

Data Index: 950123 (C4 (Web Text))

Max Activating Token Index: 162

Text #1

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.3113. Min Act: -0.0000

Data Index: 464039 (C4 (Web Text))

Max Activating Token Index: 608

Text #2

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.2130. Min Act: -0.0000

Data Index: 1172668 (C4 (Web Text))

Max Activating Token Index: 19

Text #3

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.1981. Min Act: -0.0001

Data Index: 93097 (C4 (Web Text))

Max Activating Token Index: 211

Text #4

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.2514. Min Act: -0.0000

Data Index: 1551154 (Python Code)

Max Activating Token Index: 313

Text #5

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.1747. Min Act: -0.0000

Data Index: 492003 (C4 (Web Text))

Max Activating Token Index: 578

Text #6

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.1222. Min Act: -0.0000

Data Index: 1443198 (Python Code)

Max Activating Token Index: 765

Text #7

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.0941. Min Act: -0.0000

Data Index: 1634786 (Python Code)

Max Activating Token Index: 832

Text #8

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.0704. Min Act: -0.0001

Data Index: 1057978 (C4 (Web Text))

Max Activating Token Index: 759

Text #9

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.0765. Min Act: -0.0001

Data Index: 1173211 (C4 (Web Text))

Max Activating Token Index: 642

Text #10

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.0637. Min Act: -0.0000

Data Index: 11136 (C4 (Web Text))

Max Activating Token Index: 969

Text #11

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.1158. Min Act: -0.0001

Data Index: 196672 (C4 (Web Text))

Max Activating Token Index: 914

Text #12

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.8785. Min Act: -0.0000

Data Index: 520944 (C4 (Web Text))

Max Activating Token Index: 294

Text #13

Max Range: 2.0250. Min Range: -2.0250

Max Act: 1.0030. Min Act: -0.0000

Data Index: 1383639 (Python Code)

Max Activating Token Index: 1008

Text #14

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9721. Min Act: -0.0000

Data Index: 1571179 (Python Code)

Max Activating Token Index: 550

Text #15

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9860. Min Act: -0.0000

Data Index: 1392236 (Python Code)

Max Activating Token Index: 964

Text #16

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9829. Min Act: -0.0000

Data Index: 895402 (C4 (Web Text))

Max Activating Token Index: 25

Text #17

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9630. Min Act: -0.0000

Data Index: 456768 (C4 (Web Text))

Max Activating Token Index: 893

Text #18

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9293. Min Act: -0.0000

Data Index: 94044 (C4 (Web Text))

Max Activating Token Index: 400

Text #19

Max Range: 2.0250. Min Range: -2.0250

Max Act: 0.9356. Min Act: -0.0001

Data Index: 21845 (C4 (Web Text))

Max Activating Token Index: 363