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

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

Neuron 3999 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.8907. Min Act: -0.0000

Data Index: 321862 (C4 (Web Text))

Max Activating Token Index: 724

Text #1

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.6943. Min Act: -0.0000

Data Index: 130697 (C4 (Web Text))

Max Activating Token Index: 919

Text #2

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.7869. Min Act: -0.0000

Data Index: 76105 (C4 (Web Text))

Max Activating Token Index: 373

Text #3

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.5573. Min Act: -0.0000

Data Index: 1544561 (Python Code)

Max Activating Token Index: 509

Text #4

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.5268. Min Act: -0.0000

Data Index: 1402575 (Python Code)

Max Activating Token Index: 346

Text #5

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.4484. Min Act: -0.0000

Data Index: 1399285 (Python Code)

Max Activating Token Index: 623

Text #6

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3410. Min Act: -0.0000

Data Index: 1269604 (C4 (Web Text))

Max Activating Token Index: 71

Text #7

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.4237. Min Act: -0.0000

Data Index: 1147150 (C4 (Web Text))

Max Activating Token Index: 33

Text #8

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3726. Min Act: -0.0000

Data Index: 1596963 (Python Code)

Max Activating Token Index: 936

Text #9

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2707. Min Act: -0.0000

Data Index: 1630477 (Python Code)

Max Activating Token Index: 726

Text #10

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3160. Min Act: -0.0000

Data Index: 1411270 (Python Code)

Max Activating Token Index: 260

Text #11

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2974. Min Act: -0.0000

Data Index: 1653994 (Python Code)

Max Activating Token Index: 262

Text #12

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2987. Min Act: -0.0000

Data Index: 1576888 (Python Code)

Max Activating Token Index: 151

Text #13

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3632. Min Act: -0.0000

Data Index: 1405403 (Python Code)

Max Activating Token Index: 820

Text #14

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3495. Min Act: -0.0000

Data Index: 1064778 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2637. Min Act: -0.0000

Data Index: 1492762 (Python Code)

Max Activating Token Index: 357

Text #16

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2800. Min Act: -0.0000

Data Index: 1391360 (Python Code)

Max Activating Token Index: 639

Text #17

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.2796. Min Act: -0.0000

Data Index: 1463887 (Python Code)

Max Activating Token Index: 327

Text #18

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.3380. Min Act: -0.0000

Data Index: 643211 (C4 (Web Text))

Max Activating Token Index: 471

Text #19

Max Range: 1.8907. Min Range: -1.8907

Max Act: 1.4348. Min Act: -0.0000

Data Index: 1449474 (Python Code)

Max Activating Token Index: 338