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

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

Neuron 2728 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.1186. Min Act: -0.0000

Data Index: 592909 (C4 (Web Text))

Max Activating Token Index: 935

Text #1

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.3619. Min Act: -0.0000

Data Index: 1202023 (C4 (Web Text))

Max Activating Token Index: 502

Text #2

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.3923. Min Act: -0.0000

Data Index: 1096533 (C4 (Web Text))

Max Activating Token Index: 832

Text #3

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.2449. Min Act: -0.0000

Data Index: 665210 (C4 (Web Text))

Max Activating Token Index: 557

Text #4

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.2405. Min Act: -0.0000

Data Index: 604539 (C4 (Web Text))

Max Activating Token Index: 740

Text #5

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.1351. Min Act: -0.0000

Data Index: 1645994 (Python Code)

Max Activating Token Index: 983

Text #6

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.9815. Min Act: -0.0000

Data Index: 1388504 (Python Code)

Max Activating Token Index: 10

Text #7

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.8616. Min Act: -0.0000

Data Index: 1606186 (Python Code)

Max Activating Token Index: 598

Text #8

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7025. Min Act: -0.0000

Data Index: 151498 (C4 (Web Text))

Max Activating Token Index: 802

Text #9

Max Range: 2.3923. Min Range: -2.3923

Max Act: 2.0067. Min Act: -0.0000

Data Index: 1375988 (Python Code)

Max Activating Token Index: 235

Text #10

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.6388. Min Act: -0.0000

Data Index: 849718 (C4 (Web Text))

Max Activating Token Index: 139

Text #11

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7761. Min Act: -0.0000

Data Index: 423492 (C4 (Web Text))

Max Activating Token Index: 89

Text #12

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.6179. Min Act: -0.0000

Data Index: 272844 (C4 (Web Text))

Max Activating Token Index: 732

Text #13

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7719. Min Act: -0.0000

Data Index: 1582200 (Python Code)

Max Activating Token Index: 93

Text #14

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.8709. Min Act: -0.0000

Data Index: 1428257 (Python Code)

Max Activating Token Index: 836

Text #15

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7159. Min Act: -0.0000

Data Index: 1538985 (Python Code)

Max Activating Token Index: 574

Text #16

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.6250. Min Act: -0.0000

Data Index: 1309318 (C4 (Web Text))

Max Activating Token Index: 311

Text #17

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7944. Min Act: -0.0000

Data Index: 1453920 (Python Code)

Max Activating Token Index: 40

Text #18

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.5573. Min Act: -0.0000

Data Index: 217660 (C4 (Web Text))

Max Activating Token Index: 832

Text #19

Max Range: 2.3923. Min Range: -2.3923

Max Act: 1.7190. Min Act: -0.0000

Data Index: 1161414 (C4 (Web Text))

Max Activating Token Index: 803