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

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

Neuron 5394 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.4773. Min Act: -0.0002

Data Index: 203971 (C4 (Web Text))

Max Activating Token Index: 353

Text #1

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.1798. Min Act: -0.0002

Data Index: 1041953 (C4 (Web Text))

Max Activating Token Index: 8

Text #2

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.1735. Min Act: -0.0002

Data Index: 814777 (C4 (Web Text))

Max Activating Token Index: 16

Text #3

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.0754. Min Act: -0.0002

Data Index: 1142816 (C4 (Web Text))

Max Activating Token Index: 279

Text #4

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.0942. Min Act: -0.0002

Data Index: 31085 (C4 (Web Text))

Max Activating Token Index: 596

Text #5

Max Range: 1.4773. Min Range: -1.4773

Max Act: 1.0144. Min Act: -0.0002

Data Index: 196602 (C4 (Web Text))

Max Activating Token Index: 80

Text #6

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.9073. Min Act: -0.0002

Data Index: 265281 (C4 (Web Text))

Max Activating Token Index: 732

Text #7

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.9229. Min Act: -0.0002

Data Index: 292476 (C4 (Web Text))

Max Activating Token Index: 722

Text #8

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8650. Min Act: -0.0001

Data Index: 90360 (C4 (Web Text))

Max Activating Token Index: 143

Text #9

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8825. Min Act: -0.0002

Data Index: 698090 (C4 (Web Text))

Max Activating Token Index: 820

Text #10

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8596. Min Act: -0.0002

Data Index: 1017061 (C4 (Web Text))

Max Activating Token Index: 241

Text #11

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.9426. Min Act: -0.0002

Data Index: 1082716 (C4 (Web Text))

Max Activating Token Index: 256

Text #12

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8704. Min Act: -0.0002

Data Index: 1625208 (Python Code)

Max Activating Token Index: 445

Text #13

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8307. Min Act: -0.0002

Data Index: 1293142 (C4 (Web Text))

Max Activating Token Index: 215

Text #14

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8347. Min Act: -0.0002

Data Index: 286054 (C4 (Web Text))

Max Activating Token Index: 64

Text #15

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.7552. Min Act: -0.0001

Data Index: 1306877 (C4 (Web Text))

Max Activating Token Index: 231

Text #16

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8491. Min Act: -0.0002

Data Index: 933272 (C4 (Web Text))

Max Activating Token Index: 500

Text #17

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8118. Min Act: -0.0002

Data Index: 361702 (C4 (Web Text))

Max Activating Token Index: 1011

Text #18

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.7819. Min Act: -0.0002

Data Index: 764146 (C4 (Web Text))

Max Activating Token Index: 495

Text #19

Max Range: 1.4773. Min Range: -1.4773

Max Act: 0.8020. Min Act: -0.0002

Data Index: 9325 (C4 (Web Text))

Max Activating Token Index: 190