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

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

Neuron 4689 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: 4.4853. Min Range: -4.4853

Max Act: 4.4853. Min Act: -0.0000

Data Index: 1430706 (Python Code)

Max Activating Token Index: 815

Text #1

Max Range: 4.4853. Min Range: -4.4853

Max Act: 4.3855. Min Act: -0.0000

Data Index: 1017944 (C4 (Web Text))

Max Activating Token Index: 198

Text #2

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.9297. Min Act: -0.0000

Data Index: 591622 (C4 (Web Text))

Max Activating Token Index: 678

Text #3

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.5618. Min Act: -0.0000

Data Index: 1549441 (Python Code)

Max Activating Token Index: 18

Text #4

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.4884. Min Act: -0.0000

Data Index: 1098958 (C4 (Web Text))

Max Activating Token Index: 821

Text #5

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.2945. Min Act: -0.0000

Data Index: 1608057 (Python Code)

Max Activating Token Index: 998

Text #6

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.5441. Min Act: -0.0000

Data Index: 1577843 (Python Code)

Max Activating Token Index: 120

Text #7

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.5343. Min Act: -0.0000

Data Index: 1561947 (Python Code)

Max Activating Token Index: 483

Text #8

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.5501. Min Act: -0.0000

Data Index: 1603476 (Python Code)

Max Activating Token Index: 457

Text #9

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.6986. Min Act: -0.0000

Data Index: 35647 (C4 (Web Text))

Max Activating Token Index: 415

Text #10

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.3604. Min Act: -0.0000

Data Index: 477787 (C4 (Web Text))

Max Activating Token Index: 202

Text #11

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.3018. Min Act: -0.0000

Data Index: 147771 (C4 (Web Text))

Max Activating Token Index: 438

Text #12

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.4883. Min Act: -0.0000

Data Index: 645133 (C4 (Web Text))

Max Activating Token Index: 129

Text #13

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.1257. Min Act: -0.0000

Data Index: 1532766 (Python Code)

Max Activating Token Index: 160

Text #14

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.2712. Min Act: -0.0000

Data Index: 1009283 (C4 (Web Text))

Max Activating Token Index: 387

Text #15

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.4188. Min Act: -0.0000

Data Index: 1219033 (C4 (Web Text))

Max Activating Token Index: 184

Text #16

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.4134. Min Act: -0.0000

Data Index: 1124247 (C4 (Web Text))

Max Activating Token Index: 341

Text #17

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.2202. Min Act: -0.0000

Data Index: 1492045 (Python Code)

Max Activating Token Index: 159

Text #18

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.2079. Min Act: -0.0000

Data Index: 1516652 (Python Code)

Max Activating Token Index: 871

Text #19

Max Range: 4.4853. Min Range: -4.4853

Max Act: 3.1960. Min Act: -0.0000

Data Index: 1507935 (Python Code)

Max Activating Token Index: 58