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

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

Neuron 4251 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: 0.8308. Min Range: -0.8308

Max Act: 0.8308. Min Act: -0.0002

Data Index: 552977 (C4 (Web Text))

Max Activating Token Index: 459

Text #1

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.7157. Min Act: -0.0002

Data Index: 1300381 (C4 (Web Text))

Max Activating Token Index: 970

Text #2

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.7191. Min Act: -0.0002

Data Index: 265933 (C4 (Web Text))

Max Activating Token Index: 1002

Text #3

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.7008. Min Act: -0.0002

Data Index: 756205 (C4 (Web Text))

Max Activating Token Index: 617

Text #4

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6872. Min Act: -0.0002

Data Index: 833197 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6864. Min Act: -0.0002

Data Index: 1352919 (C4 (Web Text))

Max Activating Token Index: 701

Text #6

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6802. Min Act: -0.0002

Data Index: 1147397 (C4 (Web Text))

Max Activating Token Index: 320

Text #7

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6730. Min Act: -0.0002

Data Index: 1047351 (C4 (Web Text))

Max Activating Token Index: 691

Text #8

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6630. Min Act: -0.0002

Data Index: 1587034 (Python Code)

Max Activating Token Index: 987

Text #9

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6661. Min Act: -0.0002

Data Index: 1617087 (Python Code)

Max Activating Token Index: 717

Text #10

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6586. Min Act: -0.0002

Data Index: 1244304 (C4 (Web Text))

Max Activating Token Index: 708

Text #11

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6575. Min Act: -0.0002

Data Index: 300644 (C4 (Web Text))

Max Activating Token Index: 228

Text #12

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6585. Min Act: -0.0002

Data Index: 895959 (C4 (Web Text))

Max Activating Token Index: 815

Text #13

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6412. Min Act: -0.0002

Data Index: 926989 (C4 (Web Text))

Max Activating Token Index: 819

Text #14

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6400. Min Act: -0.0002

Data Index: 328161 (C4 (Web Text))

Max Activating Token Index: 885

Text #15

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6350. Min Act: -0.0002

Data Index: 881468 (C4 (Web Text))

Max Activating Token Index: 649

Text #16

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6366. Min Act: -0.0002

Data Index: 1284008 (C4 (Web Text))

Max Activating Token Index: 9

Text #17

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6435. Min Act: -0.0002

Data Index: 107850 (C4 (Web Text))

Max Activating Token Index: 963

Text #18

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6414. Min Act: -0.0002

Data Index: 1361958 (Python Code)

Max Activating Token Index: 546

Text #19

Max Range: 0.8308. Min Range: -0.8308

Max Act: 0.6346. Min Act: -0.0002

Data Index: 984589 (C4 (Web Text))

Max Activating Token Index: 536