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

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

Neuron 535 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.6398. Min Act: -0.0000

Data Index: 98645 (C4 (Web Text))

Max Activating Token Index: 602

Text #1

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.7981. Min Act: -0.0000

Data Index: 126156 (C4 (Web Text))

Max Activating Token Index: 326

Text #2

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.4883. Min Act: -0.0000

Data Index: 56108 (C4 (Web Text))

Max Activating Token Index: 423

Text #3

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.4778. Min Act: -0.0000

Data Index: 386690 (C4 (Web Text))

Max Activating Token Index: 580

Text #4

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2111. Min Act: -0.0000

Data Index: 903852 (C4 (Web Text))

Max Activating Token Index: 426

Text #5

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.0791. Min Act: -0.0000

Data Index: 231399 (C4 (Web Text))

Max Activating Token Index: 193

Text #6

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2905. Min Act: -0.0000

Data Index: 1191585 (C4 (Web Text))

Max Activating Token Index: 198

Text #7

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2411. Min Act: -0.0000

Data Index: 746378 (C4 (Web Text))

Max Activating Token Index: 137

Text #8

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.5378. Min Act: -0.0000

Data Index: 1226671 (C4 (Web Text))

Max Activating Token Index: 418

Text #9

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.1520. Min Act: -0.0000

Data Index: 985350 (C4 (Web Text))

Max Activating Token Index: 133

Text #10

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2466. Min Act: -0.0000

Data Index: 501025 (C4 (Web Text))

Max Activating Token Index: 71

Text #11

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.1170. Min Act: -0.0000

Data Index: 8376 (C4 (Web Text))

Max Activating Token Index: 12

Text #12

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2291. Min Act: -0.0000

Data Index: 254644 (C4 (Web Text))

Max Activating Token Index: 230

Text #13

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.0870. Min Act: -0.0000

Data Index: 270360 (C4 (Web Text))

Max Activating Token Index: 286

Text #14

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.1083. Min Act: -0.0000

Data Index: 945625 (C4 (Web Text))

Max Activating Token Index: 36

Text #15

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.0624. Min Act: -0.0000

Data Index: 1180099 (C4 (Web Text))

Max Activating Token Index: 43

Text #16

Max Range: 2.7981. Min Range: -2.7981

Max Act: 1.9771. Min Act: -0.0000

Data Index: 1336990 (C4 (Web Text))

Max Activating Token Index: 21

Text #17

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.1653. Min Act: -0.0000

Data Index: 481227 (C4 (Web Text))

Max Activating Token Index: 560

Text #18

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.1367. Min Act: -0.0000

Data Index: 596607 (C4 (Web Text))

Max Activating Token Index: 492

Text #19

Max Range: 2.7981. Min Range: -2.7981

Max Act: 2.2389. Min Act: -0.0000

Data Index: 702160 (C4 (Web Text))

Max Activating Token Index: 178