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

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

Neuron 3484 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: 0.8784. Min Range: -0.8784

Max Act: 0.8784. Min Act: -0.0001

Data Index: 817957 (C4 (Web Text))

Max Activating Token Index: 594

Text #1

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.8276. Min Act: -0.0001

Data Index: 888456 (C4 (Web Text))

Max Activating Token Index: 940

Text #2

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.8112. Min Act: -0.0001

Data Index: 264739 (C4 (Web Text))

Max Activating Token Index: 718

Text #3

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7788. Min Act: -0.0001

Data Index: 1574993 (Python Code)

Max Activating Token Index: 979

Text #4

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7379. Min Act: -0.0001

Data Index: 1410978 (Python Code)

Max Activating Token Index: 704

Text #5

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7465. Min Act: -0.0001

Data Index: 113470 (C4 (Web Text))

Max Activating Token Index: 262

Text #6

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7261. Min Act: -0.0001

Data Index: 553376 (C4 (Web Text))

Max Activating Token Index: 324

Text #7

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7448. Min Act: -0.0001

Data Index: 269580 (C4 (Web Text))

Max Activating Token Index: 710

Text #8

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7458. Min Act: -0.0001

Data Index: 216092 (C4 (Web Text))

Max Activating Token Index: 470

Text #9

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7539. Min Act: -0.0001

Data Index: 521370 (C4 (Web Text))

Max Activating Token Index: 746

Text #10

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7565. Min Act: -0.0001

Data Index: 47225 (C4 (Web Text))

Max Activating Token Index: 216

Text #11

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7043. Min Act: -0.0001

Data Index: 970299 (C4 (Web Text))

Max Activating Token Index: 781

Text #12

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6691. Min Act: -0.0001

Data Index: 128248 (C4 (Web Text))

Max Activating Token Index: 555

Text #13

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.7027. Min Act: -0.0001

Data Index: 272392 (C4 (Web Text))

Max Activating Token Index: 815

Text #14

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6596. Min Act: -0.0001

Data Index: 211410 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6732. Min Act: -0.0001

Data Index: 1404885 (Python Code)

Max Activating Token Index: 130

Text #16

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6146. Min Act: -0.0001

Data Index: 1076647 (C4 (Web Text))

Max Activating Token Index: 189

Text #17

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6918. Min Act: -0.0001

Data Index: 1399809 (Python Code)

Max Activating Token Index: 764

Text #18

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6558. Min Act: -0.0001

Data Index: 437173 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 0.8784. Min Range: -0.8784

Max Act: 0.6112. Min Act: -0.0001

Data Index: 1539036 (Python Code)

Max Activating Token Index: 393