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

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

Neuron 4796 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.4614. Min Act: -0.0001

Data Index: 1450085 (Python Code)

Max Activating Token Index: 585

Text #1

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.4585. Min Act: -0.0000

Data Index: 1506701 (Python Code)

Max Activating Token Index: 327

Text #2

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.4045. Min Act: -0.0000

Data Index: 1450952 (Python Code)

Max Activating Token Index: 190

Text #3

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.1923. Min Act: -0.0000

Data Index: 1429366 (Python Code)

Max Activating Token Index: 851

Text #4

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.3137. Min Act: -0.0000

Data Index: 1449575 (Python Code)

Max Activating Token Index: 814

Text #5

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.2267. Min Act: -0.0000

Data Index: 1309017 (C4 (Web Text))

Max Activating Token Index: 374

Text #6

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.2322. Min Act: -0.0000

Data Index: 1639963 (Python Code)

Max Activating Token Index: 473

Text #7

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.2251. Min Act: -0.0000

Data Index: 1411133 (Python Code)

Max Activating Token Index: 240

Text #8

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.2117. Min Act: -0.0000

Data Index: 1627009 (Python Code)

Max Activating Token Index: 585

Text #9

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.9543. Min Act: -0.0000

Data Index: 1469565 (Python Code)

Max Activating Token Index: 274

Text #10

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.8463. Min Act: -0.0000

Data Index: 1424103 (Python Code)

Max Activating Token Index: 59

Text #11

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.8945. Min Act: -0.0000

Data Index: 528847 (C4 (Web Text))

Max Activating Token Index: 98

Text #12

Max Range: 2.4614. Min Range: -2.4614

Max Act: 2.0836. Min Act: -0.0000

Data Index: 1455007 (Python Code)

Max Activating Token Index: 805

Text #13

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.9581. Min Act: -0.0000

Data Index: 1441951 (Python Code)

Max Activating Token Index: 157

Text #14

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.9430. Min Act: -0.0000

Data Index: 510828 (C4 (Web Text))

Max Activating Token Index: 531

Text #15

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.8252. Min Act: -0.0000

Data Index: 658612 (C4 (Web Text))

Max Activating Token Index: 52

Text #16

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.8032. Min Act: -0.0000

Data Index: 1362266 (Python Code)

Max Activating Token Index: 927

Text #17

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.7787. Min Act: -0.0000

Data Index: 1559959 (Python Code)

Max Activating Token Index: 335

Text #18

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.7526. Min Act: -0.0000

Data Index: 1418029 (Python Code)

Max Activating Token Index: 283

Text #19

Max Range: 2.4614. Min Range: -2.4614

Max Act: 1.7589. Min Act: -0.0001

Data Index: 1385229 (Python Code)

Max Activating Token Index: 263