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

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

Neuron 2080 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1136. Min Range: -5.1136

Max Act: 5.1136. Min Act: -0.0000

Data Index: 648643 (C4 (Web Text))

Max Activating Token Index: 665

Text #1

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.7651. Min Act: -0.0000

Data Index: 1612731 (Python Code)

Max Activating Token Index: 786

Text #2

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.5020. Min Act: -0.0000

Data Index: 1470330 (Python Code)

Max Activating Token Index: 138

Text #3

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.3876. Min Act: -0.0000

Data Index: 1546527 (Python Code)

Max Activating Token Index: 156

Text #4

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.3042. Min Act: -0.0000

Data Index: 1405893 (Python Code)

Max Activating Token Index: 686

Text #5

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.3082. Min Act: -0.0000

Data Index: 1487876 (Python Code)

Max Activating Token Index: 495

Text #6

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.9108. Min Act: -0.0000

Data Index: 1339433 (C4 (Web Text))

Max Activating Token Index: 295

Text #7

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.2030. Min Act: -0.0000

Data Index: 1377320 (Python Code)

Max Activating Token Index: 737

Text #8

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.8648. Min Act: -0.0000

Data Index: 1330921 (C4 (Web Text))

Max Activating Token Index: 747

Text #9

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.1377. Min Act: -0.0000

Data Index: 686652 (C4 (Web Text))

Max Activating Token Index: 679

Text #10

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.3287. Min Act: -0.0000

Data Index: 188525 (C4 (Web Text))

Max Activating Token Index: 189

Text #11

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.7964. Min Act: -0.0000

Data Index: 970708 (C4 (Web Text))

Max Activating Token Index: 788

Text #12

Max Range: 5.1136. Min Range: -5.1136

Max Act: 4.1161. Min Act: -0.0000

Data Index: 181477 (C4 (Web Text))

Max Activating Token Index: 902

Text #13

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.9413. Min Act: -0.0000

Data Index: 1495288 (Python Code)

Max Activating Token Index: 530

Text #14

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.9043. Min Act: -0.0000

Data Index: 1408339 (Python Code)

Max Activating Token Index: 320

Text #15

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.9002. Min Act: -0.0000

Data Index: 1565447 (Python Code)

Max Activating Token Index: 411

Text #16

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.8589. Min Act: -0.0000

Data Index: 1362242 (Python Code)

Max Activating Token Index: 839

Text #17

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.8554. Min Act: -0.0000

Data Index: 238355 (C4 (Web Text))

Max Activating Token Index: 369

Text #18

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.6931. Min Act: -0.0000

Data Index: 1547768 (Python Code)

Max Activating Token Index: 419

Text #19

Max Range: 5.1136. Min Range: -5.1136

Max Act: 3.6662. Min Act: -0.0000

Data Index: 1388969 (Python Code)

Max Activating Token Index: 810