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

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

Neuron 640 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: 4.2965. Min Range: -4.2965

Max Act: 4.2965. Min Act: -0.0000

Data Index: 1600349 (Python Code)

Max Activating Token Index: 654

Text #1

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.9277. Min Act: -0.0000

Data Index: 1528107 (Python Code)

Max Activating Token Index: 405

Text #2

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.6328. Min Act: -0.0000

Data Index: 1389415 (Python Code)

Max Activating Token Index: 671

Text #3

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.5656. Min Act: -0.0000

Data Index: 440680 (C4 (Web Text))

Max Activating Token Index: 604

Text #4

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.6533. Min Act: -0.0000

Data Index: 1394028 (Python Code)

Max Activating Token Index: 955

Text #5

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.4534. Min Act: -0.0000

Data Index: 1640090 (Python Code)

Max Activating Token Index: 211

Text #6

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.2728. Min Act: -0.0000

Data Index: 21388 (C4 (Web Text))

Max Activating Token Index: 970

Text #7

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.1974. Min Act: -0.0000

Data Index: 1024646 (C4 (Web Text))

Max Activating Token Index: 880

Text #8

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.0829. Min Act: -0.0000

Data Index: 1638638 (Python Code)

Max Activating Token Index: 695

Text #9

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.0649. Min Act: -0.0000

Data Index: 998026 (C4 (Web Text))

Max Activating Token Index: 332

Text #10

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.8946. Min Act: -0.0000

Data Index: 1652563 (Python Code)

Max Activating Token Index: 214

Text #11

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.9894. Min Act: -0.0000

Data Index: 513639 (C4 (Web Text))

Max Activating Token Index: 259

Text #12

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.9314. Min Act: -0.0000

Data Index: 1441715 (Python Code)

Max Activating Token Index: 592

Text #13

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.9001. Min Act: -0.0000

Data Index: 1054098 (C4 (Web Text))

Max Activating Token Index: 492

Text #14

Max Range: 4.2965. Min Range: -4.2965

Max Act: 3.0296. Min Act: -0.0000

Data Index: 497755 (C4 (Web Text))

Max Activating Token Index: 191

Text #15

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.8717. Min Act: -0.0000

Data Index: 94791 (C4 (Web Text))

Max Activating Token Index: 496

Text #16

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.6941. Min Act: -0.0000

Data Index: 1165843 (C4 (Web Text))

Max Activating Token Index: 335

Text #17

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.7869. Min Act: -0.0000

Data Index: 1311707 (C4 (Web Text))

Max Activating Token Index: 590

Text #18

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.6504. Min Act: -0.0000

Data Index: 1379123 (Python Code)

Max Activating Token Index: 732

Text #19

Max Range: 4.2965. Min Range: -4.2965

Max Act: 2.8945. Min Act: -0.0000

Data Index: 313948 (C4 (Web Text))

Max Activating Token Index: 807