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

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

Neuron 1605 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.7902. Min Range: -4.7902

Max Act: 4.7902. Min Act: -0.0000

Data Index: 232485 (C4 (Web Text))

Max Activating Token Index: 65

Text #1

Max Range: 4.7902. Min Range: -4.7902

Max Act: 4.3411. Min Act: -0.0000

Data Index: 1194884 (C4 (Web Text))

Max Activating Token Index: 181

Text #2

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.9074. Min Act: -0.0000

Data Index: 641093 (C4 (Web Text))

Max Activating Token Index: 182

Text #3

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.4649. Min Act: -0.0000

Data Index: 1470752 (Python Code)

Max Activating Token Index: 201

Text #4

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.2338. Min Act: -0.0000

Data Index: 1422690 (Python Code)

Max Activating Token Index: 409

Text #5

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.4038. Min Act: -0.0000

Data Index: 176195 (C4 (Web Text))

Max Activating Token Index: 495

Text #6

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.2280. Min Act: -0.0000

Data Index: 1546666 (Python Code)

Max Activating Token Index: 55

Text #7

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.1787. Min Act: -0.0000

Data Index: 1419521 (Python Code)

Max Activating Token Index: 483

Text #8

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.0929. Min Act: -0.0000

Data Index: 310909 (C4 (Web Text))

Max Activating Token Index: 54

Text #9

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.0650. Min Act: -0.0000

Data Index: 1308556 (C4 (Web Text))

Max Activating Token Index: 52

Text #10

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.7320. Min Act: -0.0000

Data Index: 587222 (C4 (Web Text))

Max Activating Token Index: 420

Text #11

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.0373. Min Act: -0.0000

Data Index: 447676 (C4 (Web Text))

Max Activating Token Index: 447

Text #12

Max Range: 4.7902. Min Range: -4.7902

Max Act: 3.1565. Min Act: -0.0000

Data Index: 435949 (C4 (Web Text))

Max Activating Token Index: 334

Text #13

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.9840. Min Act: -0.0000

Data Index: 1454552 (Python Code)

Max Activating Token Index: 64

Text #14

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.9668. Min Act: -0.0000

Data Index: 1366974 (Python Code)

Max Activating Token Index: 339

Text #15

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.9741. Min Act: -0.0000

Data Index: 198144 (C4 (Web Text))

Max Activating Token Index: 801

Text #16

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.9451. Min Act: -0.0000

Data Index: 1583128 (Python Code)

Max Activating Token Index: 709

Text #17

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.9259. Min Act: -0.0000

Data Index: 1380617 (Python Code)

Max Activating Token Index: 405

Text #18

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.8080. Min Act: -0.0000

Data Index: 1572081 (Python Code)

Max Activating Token Index: 113

Text #19

Max Range: 4.7902. Min Range: -4.7902

Max Act: 2.7690. Min Act: -0.0000

Data Index: 1422078 (Python Code)

Max Activating Token Index: 47