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

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

Neuron 556 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8940. Min Act: -0.0001

Data Index: 437037 (C4 (Web Text))

Max Activating Token Index: 600

Text #1

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8468. Min Act: -0.0001

Data Index: 620869 (C4 (Web Text))

Max Activating Token Index: 132

Text #2

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.9080. Min Act: -0.0001

Data Index: 429928 (C4 (Web Text))

Max Activating Token Index: 806

Text #3

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8663. Min Act: -0.0001

Data Index: 492459 (C4 (Web Text))

Max Activating Token Index: 154

Text #4

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8932. Min Act: -0.0001

Data Index: 575328 (C4 (Web Text))

Max Activating Token Index: 286

Text #5

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.7570. Min Act: -0.0001

Data Index: 1200669 (C4 (Web Text))

Max Activating Token Index: 215

Text #6

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8544. Min Act: -0.0001

Data Index: 84958 (C4 (Web Text))

Max Activating Token Index: 167

Text #7

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8260. Min Act: -0.0001

Data Index: 678282 (C4 (Web Text))

Max Activating Token Index: 180

Text #8

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8656. Min Act: -0.0001

Data Index: 948968 (C4 (Web Text))

Max Activating Token Index: 338

Text #9

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8900. Min Act: -0.0001

Data Index: 126600 (C4 (Web Text))

Max Activating Token Index: 247

Text #10

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8417. Min Act: -0.0001

Data Index: 81189 (C4 (Web Text))

Max Activating Token Index: 181

Text #11

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8522. Min Act: -0.0001

Data Index: 1176219 (C4 (Web Text))

Max Activating Token Index: 504

Text #12

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8490. Min Act: -0.0001

Data Index: 788587 (C4 (Web Text))

Max Activating Token Index: 700

Text #13

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.7079. Min Act: -0.0001

Data Index: 585140 (C4 (Web Text))

Max Activating Token Index: 157

Text #14

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8072. Min Act: -0.0001

Data Index: 940304 (C4 (Web Text))

Max Activating Token Index: 523

Text #15

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8373. Min Act: -0.0001

Data Index: 940639 (C4 (Web Text))

Max Activating Token Index: 203

Text #16

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.7761. Min Act: -0.0001

Data Index: 957260 (C4 (Web Text))

Max Activating Token Index: 56

Text #17

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8331. Min Act: -0.0001

Data Index: 466982 (C4 (Web Text))

Max Activating Token Index: 223

Text #18

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.8012. Min Act: -0.0001

Data Index: 593937 (C4 (Web Text))

Max Activating Token Index: 416

Text #19

Max Range: 0.9080. Min Range: -0.9080

Max Act: 0.7777. Min Act: -0.0001

Data Index: 391200 (C4 (Web Text))

Max Activating Token Index: 31