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

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

Neuron 2628 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.8458. Min Act: -0.0000

Data Index: 310390 (C4 (Web Text))

Max Activating Token Index: 780

Text #1

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.4301. Min Act: -0.0000

Data Index: 1643254 (Python Code)

Max Activating Token Index: 141

Text #2

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.5666. Min Act: -0.0000

Data Index: 301260 (C4 (Web Text))

Max Activating Token Index: 574

Text #3

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.5624. Min Act: -0.0000

Data Index: 513871 (C4 (Web Text))

Max Activating Token Index: 31

Text #4

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.5429. Min Act: -0.0000

Data Index: 648739 (C4 (Web Text))

Max Activating Token Index: 701

Text #5

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.6747. Min Act: -0.0000

Data Index: 974921 (C4 (Web Text))

Max Activating Token Index: 959

Text #6

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.2536. Min Act: -0.0000

Data Index: 439342 (C4 (Web Text))

Max Activating Token Index: 591

Text #7

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.2211. Min Act: -0.0000

Data Index: 1607499 (Python Code)

Max Activating Token Index: 871

Text #8

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.1227. Min Act: -0.0000

Data Index: 376774 (C4 (Web Text))

Max Activating Token Index: 393

Text #9

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.9154. Min Act: -0.0000

Data Index: 519820 (C4 (Web Text))

Max Activating Token Index: 852

Text #10

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.0758. Min Act: -0.0000

Data Index: 1355028 (C4 (Web Text))

Max Activating Token Index: 216

Text #11

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.0578. Min Act: -0.0000

Data Index: 202707 (C4 (Web Text))

Max Activating Token Index: 185

Text #12

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.0666. Min Act: -0.0000

Data Index: 1011869 (C4 (Web Text))

Max Activating Token Index: 328

Text #13

Max Range: 3.8458. Min Range: -3.8458

Max Act: 3.0218. Min Act: -0.0000

Data Index: 1367467 (Python Code)

Max Activating Token Index: 1021

Text #14

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.8498. Min Act: -0.0000

Data Index: 1568175 (Python Code)

Max Activating Token Index: 601

Text #15

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.8047. Min Act: -0.0000

Data Index: 1069841 (C4 (Web Text))

Max Activating Token Index: 60

Text #16

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.8743. Min Act: -0.0000

Data Index: 1030700 (C4 (Web Text))

Max Activating Token Index: 372

Text #17

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.7157. Min Act: -0.0000

Data Index: 1634686 (Python Code)

Max Activating Token Index: 480

Text #18

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.5183. Min Act: -0.0000

Data Index: 1416416 (Python Code)

Max Activating Token Index: 374

Text #19

Max Range: 3.8458. Min Range: -3.8458

Max Act: 2.6700. Min Act: -0.0000

Data Index: 505953 (C4 (Web Text))

Max Activating Token Index: 260