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

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

Neuron 895 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.6224. Min Act: -0.0002

Data Index: 932894 (C4 (Web Text))

Max Activating Token Index: 2

Text #1

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.6224. Min Act: -0.0002

Data Index: 1452521 (Python Code)

Max Activating Token Index: 2

Text #2

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.6224. Min Act: -0.0002

Data Index: 608750 (C4 (Web Text))

Max Activating Token Index: 2

Text #3

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.6687. Min Act: -0.0002

Data Index: 1447954 (Python Code)

Max Activating Token Index: 3

Text #4

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.6555. Min Act: -0.0002

Data Index: 335685 (C4 (Web Text))

Max Activating Token Index: 709

Text #5

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.4844. Min Act: -0.0002

Data Index: 1582972 (Python Code)

Max Activating Token Index: 749

Text #6

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3967. Min Act: -0.0002

Data Index: 798569 (C4 (Web Text))

Max Activating Token Index: 182

Text #7

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3728. Min Act: -0.0002

Data Index: 237888 (C4 (Web Text))

Max Activating Token Index: 708

Text #8

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3840. Min Act: -0.0002

Data Index: 1482750 (Python Code)

Max Activating Token Index: 616

Text #9

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3639. Min Act: -0.0002

Data Index: 1110207 (C4 (Web Text))

Max Activating Token Index: 764

Text #10

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3349. Min Act: -0.0002

Data Index: 1515143 (Python Code)

Max Activating Token Index: 28

Text #11

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3136. Min Act: -0.0002

Data Index: 1464695 (Python Code)

Max Activating Token Index: 666

Text #12

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3288. Min Act: -0.0002

Data Index: 1048205 (C4 (Web Text))

Max Activating Token Index: 367

Text #13

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3270. Min Act: -0.0002

Data Index: 1529273 (Python Code)

Max Activating Token Index: 782

Text #14

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3214. Min Act: -0.0002

Data Index: 1648417 (Python Code)

Max Activating Token Index: 664

Text #15

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.3059. Min Act: -0.0002

Data Index: 333041 (C4 (Web Text))

Max Activating Token Index: 902

Text #16

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.2625. Min Act: -0.0002

Data Index: 1530150 (Python Code)

Max Activating Token Index: 909

Text #17

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.2750. Min Act: -0.0002

Data Index: 1116265 (C4 (Web Text))

Max Activating Token Index: 11

Text #18

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.2749. Min Act: -0.0002

Data Index: 1101445 (C4 (Web Text))

Max Activating Token Index: 265

Text #19

Max Range: 1.6687. Min Range: -1.6687

Max Act: 1.2716. Min Act: -0.0002

Data Index: 1657061 (Python Code)

Max Activating Token Index: 763