Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3128 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.6396. Min Act: -0.0001

Data Index: 844233 (C4 (Web Text))

Max Activating Token Index: 83

Text #1

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.6279. Min Act: -0.0001

Data Index: 38833 (C4 (Web Text))

Max Activating Token Index: 672

Text #2

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5941. Min Act: -0.0001

Data Index: 1300752 (C4 (Web Text))

Max Activating Token Index: 267

Text #3

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.6135. Min Act: -0.0001

Data Index: 1021582 (C4 (Web Text))

Max Activating Token Index: 694

Text #4

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5944. Min Act: -0.0001

Data Index: 52385 (C4 (Web Text))

Max Activating Token Index: 385

Text #5

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.6114. Min Act: -0.0001

Data Index: 592460 (C4 (Web Text))

Max Activating Token Index: 61

Text #6

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5930. Min Act: -0.0001

Data Index: 336228 (C4 (Web Text))

Max Activating Token Index: 368

Text #7

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5815. Min Act: -0.0001

Data Index: 172680 (C4 (Web Text))

Max Activating Token Index: 731

Text #8

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5608. Min Act: -0.0001

Data Index: 606874 (C4 (Web Text))

Max Activating Token Index: 129

Text #9

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5835. Min Act: -0.0001

Data Index: 322551 (C4 (Web Text))

Max Activating Token Index: 42

Text #10

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5256. Min Act: -0.0001

Data Index: 1055002 (C4 (Web Text))

Max Activating Token Index: 92

Text #11

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5564. Min Act: -0.0001

Data Index: 1535987 (Python Code)

Max Activating Token Index: 341

Text #12

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5779. Min Act: -0.0001

Data Index: 1004364 (C4 (Web Text))

Max Activating Token Index: 453

Text #13

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5684. Min Act: -0.0001

Data Index: 916848 (C4 (Web Text))

Max Activating Token Index: 554

Text #14

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5253. Min Act: -0.0001

Data Index: 1243122 (C4 (Web Text))

Max Activating Token Index: 38

Text #15

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5389. Min Act: -0.0001

Data Index: 517162 (C4 (Web Text))

Max Activating Token Index: 236

Text #16

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5329. Min Act: -0.0001

Data Index: 99092 (C4 (Web Text))

Max Activating Token Index: 83

Text #17

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5160. Min Act: -0.0001

Data Index: 1074569 (C4 (Web Text))

Max Activating Token Index: 506

Text #18

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5060. Min Act: -0.0001

Data Index: 365654 (C4 (Web Text))

Max Activating Token Index: 49

Text #19

Max Range: 0.6396. Min Range: -0.6396

Max Act: 0.5044. Min Act: -0.0001

Data Index: 90218 (C4 (Web Text))

Max Activating Token Index: 90