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

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

Neuron 4583 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: 2.4081. Min Range: -2.4081

Max Act: 2.4081. Min Act: -0.0000

Data Index: 926040 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.6264. Min Act: -0.0000

Data Index: 471704 (C4 (Web Text))

Max Activating Token Index: 701

Text #2

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.6974. Min Act: -0.0000

Data Index: 895863 (C4 (Web Text))

Max Activating Token Index: 664

Text #3

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.6730. Min Act: -0.0000

Data Index: 1382912 (Python Code)

Max Activating Token Index: 84

Text #4

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.5169. Min Act: -0.0000

Data Index: 1464738 (Python Code)

Max Activating Token Index: 515

Text #5

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4011. Min Act: -0.0000

Data Index: 1153511 (C4 (Web Text))

Max Activating Token Index: 748

Text #6

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4824. Min Act: -0.0000

Data Index: 407972 (C4 (Web Text))

Max Activating Token Index: 235

Text #7

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3612. Min Act: -0.0000

Data Index: 100302 (C4 (Web Text))

Max Activating Token Index: 424

Text #8

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3525. Min Act: -0.0000

Data Index: 122268 (C4 (Web Text))

Max Activating Token Index: 474

Text #9

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3478. Min Act: -0.0000

Data Index: 156710 (C4 (Web Text))

Max Activating Token Index: 898

Text #10

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4947. Min Act: -0.0000

Data Index: 781204 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4015. Min Act: -0.0000

Data Index: 1468874 (Python Code)

Max Activating Token Index: 425

Text #12

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4806. Min Act: -0.0000

Data Index: 1543376 (Python Code)

Max Activating Token Index: 970

Text #13

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3824. Min Act: -0.0000

Data Index: 1565720 (Python Code)

Max Activating Token Index: 387

Text #14

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3702. Min Act: -0.0000

Data Index: 1630450 (Python Code)

Max Activating Token Index: 398

Text #15

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4489. Min Act: -0.0000

Data Index: 675988 (C4 (Web Text))

Max Activating Token Index: 216

Text #16

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4303. Min Act: -0.0000

Data Index: 970571 (C4 (Web Text))

Max Activating Token Index: 724

Text #17

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.3366. Min Act: -0.0000

Data Index: 245198 (C4 (Web Text))

Max Activating Token Index: 471

Text #18

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.2674. Min Act: -0.0000

Data Index: 1157169 (C4 (Web Text))

Max Activating Token Index: 266

Text #19

Max Range: 2.4081. Min Range: -2.4081

Max Act: 1.4179. Min Act: -0.0000

Data Index: 149670 (C4 (Web Text))

Max Activating Token Index: 129