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

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

Neuron 3053 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.8596. Min Act: -0.0000

Data Index: 129068 (C4 (Web Text))

Max Activating Token Index: 444

Text #1

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7780. Min Act: -0.0000

Data Index: 1141566 (C4 (Web Text))

Max Activating Token Index: 395

Text #2

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7920. Min Act: -0.0000

Data Index: 485023 (C4 (Web Text))

Max Activating Token Index: 1003

Text #3

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7796. Min Act: -0.0000

Data Index: 276708 (C4 (Web Text))

Max Activating Token Index: 905

Text #4

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7504. Min Act: -0.0000

Data Index: 514180 (C4 (Web Text))

Max Activating Token Index: 471

Text #5

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7505. Min Act: -0.0000

Data Index: 671220 (C4 (Web Text))

Max Activating Token Index: 986

Text #6

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7257. Min Act: -0.0001

Data Index: 54916 (C4 (Web Text))

Max Activating Token Index: 557

Text #7

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7307. Min Act: -0.0000

Data Index: 353982 (C4 (Web Text))

Max Activating Token Index: 1021

Text #8

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7380. Min Act: -0.0001

Data Index: 1146906 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7006. Min Act: -0.0000

Data Index: 265406 (C4 (Web Text))

Max Activating Token Index: 337

Text #10

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7267. Min Act: -0.0001

Data Index: 1125898 (C4 (Web Text))

Max Activating Token Index: 514

Text #11

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7173. Min Act: -0.0001

Data Index: 223439 (C4 (Web Text))

Max Activating Token Index: 185

Text #12

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7048. Min Act: -0.0001

Data Index: 1150439 (C4 (Web Text))

Max Activating Token Index: 226

Text #13

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.7148. Min Act: -0.0001

Data Index: 926541 (C4 (Web Text))

Max Activating Token Index: 447

Text #14

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6802. Min Act: -0.0001

Data Index: 1637100 (Python Code)

Max Activating Token Index: 305

Text #15

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6888. Min Act: -0.0001

Data Index: 442557 (C4 (Web Text))

Max Activating Token Index: 622

Text #16

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6607. Min Act: -0.0001

Data Index: 902587 (C4 (Web Text))

Max Activating Token Index: 242

Text #17

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6853. Min Act: -0.0001

Data Index: 1240541 (C4 (Web Text))

Max Activating Token Index: 998

Text #18

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6389. Min Act: -0.0001

Data Index: 949384 (C4 (Web Text))

Max Activating Token Index: 127

Text #19

Max Range: 0.8596. Min Range: -0.8596

Max Act: 0.6535. Min Act: -0.0001

Data Index: 1339911 (C4 (Web Text))

Max Activating Token Index: 579