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

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

Neuron 2192 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.9091. Min Act: -0.0002

Data Index: 279069 (C4 (Web Text))

Max Activating Token Index: 999

Text #1

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.9119. Min Act: -0.0002

Data Index: 709727 (C4 (Web Text))

Max Activating Token Index: 620

Text #2

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.8325. Min Act: -0.0002

Data Index: 813652 (C4 (Web Text))

Max Activating Token Index: 955

Text #3

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.8130. Min Act: -0.0002

Data Index: 312560 (C4 (Web Text))

Max Activating Token Index: 147

Text #4

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7417. Min Act: -0.0002

Data Index: 735994 (C4 (Web Text))

Max Activating Token Index: 638

Text #5

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.8302. Min Act: -0.0001

Data Index: 456169 (C4 (Web Text))

Max Activating Token Index: 790

Text #6

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7117. Min Act: -0.0002

Data Index: 9306 (C4 (Web Text))

Max Activating Token Index: 631

Text #7

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7597. Min Act: -0.0002

Data Index: 527094 (C4 (Web Text))

Max Activating Token Index: 973

Text #8

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7297. Min Act: -0.0002

Data Index: 976705 (C4 (Web Text))

Max Activating Token Index: 884

Text #9

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7225. Min Act: -0.0002

Data Index: 379015 (C4 (Web Text))

Max Activating Token Index: 532

Text #10

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7564. Min Act: -0.0002

Data Index: 137876 (C4 (Web Text))

Max Activating Token Index: 889

Text #11

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7565. Min Act: -0.0002

Data Index: 229733 (C4 (Web Text))

Max Activating Token Index: 257

Text #12

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7114. Min Act: -0.0001

Data Index: 423924 (C4 (Web Text))

Max Activating Token Index: 6

Text #13

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.8052. Min Act: -0.0002

Data Index: 1257924 (C4 (Web Text))

Max Activating Token Index: 659

Text #14

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7023. Min Act: -0.0001

Data Index: 1502462 (Python Code)

Max Activating Token Index: 281

Text #15

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.6799. Min Act: -0.0002

Data Index: 1055565 (C4 (Web Text))

Max Activating Token Index: 968

Text #16

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.6618. Min Act: -0.0002

Data Index: 27445 (C4 (Web Text))

Max Activating Token Index: 22

Text #17

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.6761. Min Act: -0.0002

Data Index: 480841 (C4 (Web Text))

Max Activating Token Index: 761

Text #18

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.7131. Min Act: -0.0002

Data Index: 1353025 (C4 (Web Text))

Max Activating Token Index: 453

Text #19

Max Range: 0.9119. Min Range: -0.9119

Max Act: 0.6498. Min Act: -0.0002

Data Index: 332567 (C4 (Web Text))

Max Activating Token Index: 1015