Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3655 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.8262. Min Act: -0.0001

Data Index: 567223 (C4 (Web Text))

Max Activating Token Index: 167

Text #1

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6139. Min Act: -0.0001

Data Index: 964676 (C4 (Web Text))

Max Activating Token Index: 854

Text #2

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7458. Min Act: -0.0001

Data Index: 915095 (C4 (Web Text))

Max Activating Token Index: 204

Text #3

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7336. Min Act: -0.0001

Data Index: 1357912 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7250. Min Act: -0.0001

Data Index: 247763 (C4 (Web Text))

Max Activating Token Index: 13

Text #5

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6828. Min Act: -0.0001

Data Index: 1267984 (C4 (Web Text))

Max Activating Token Index: 1013

Text #6

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7254. Min Act: -0.0001

Data Index: 359566 (C4 (Web Text))

Max Activating Token Index: 182

Text #7

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7494. Min Act: -0.0001

Data Index: 602174 (C4 (Web Text))

Max Activating Token Index: 636

Text #8

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6855. Min Act: -0.0001

Data Index: 1255307 (C4 (Web Text))

Max Activating Token Index: 7

Text #9

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6614. Min Act: -0.0001

Data Index: 845784 (C4 (Web Text))

Max Activating Token Index: 342

Text #10

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7196. Min Act: -0.0001

Data Index: 1133131 (C4 (Web Text))

Max Activating Token Index: 983

Text #11

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6757. Min Act: -0.0001

Data Index: 614543 (C4 (Web Text))

Max Activating Token Index: 133

Text #12

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7200. Min Act: -0.0001

Data Index: 1079681 (C4 (Web Text))

Max Activating Token Index: 72

Text #13

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.7168. Min Act: -0.0001

Data Index: 135222 (C4 (Web Text))

Max Activating Token Index: 348

Text #14

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6668. Min Act: -0.0001

Data Index: 292739 (C4 (Web Text))

Max Activating Token Index: 406

Text #15

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6602. Min Act: -0.0001

Data Index: 520895 (C4 (Web Text))

Max Activating Token Index: 139

Text #16

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6674. Min Act: -0.0001

Data Index: 1646028 (Python Code)

Max Activating Token Index: 868

Text #17

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6854. Min Act: -0.0001

Data Index: 827045 (C4 (Web Text))

Max Activating Token Index: 872

Text #18

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6920. Min Act: -0.0001

Data Index: 759140 (C4 (Web Text))

Max Activating Token Index: 770

Text #19

Max Range: 0.8262. Min Range: -0.8262

Max Act: 0.6817. Min Act: -0.0001

Data Index: 1126486 (C4 (Web Text))

Max Activating Token Index: 113