Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 2871 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 5.1797. Min Range: -5.1797

Max Act: 5.1797. Min Act: -0.0000

Data Index: 1634803 (The Pile)

Max Activating Token Index: 402

Text #1

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.7445. Min Act: -0.0000

Data Index: 434635 (The Pile)

Max Activating Token Index: 448

Text #2

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.6164. Min Act: -0.0000

Data Index: 179610 (The Pile)

Max Activating Token Index: 732

Text #3

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.5267. Min Act: -0.0000

Data Index: 1812338 (The Pile)

Max Activating Token Index: 154

Text #4

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.4084. Min Act: -0.0000

Data Index: 1971099 (The Pile)

Max Activating Token Index: 512

Text #5

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.2240. Min Act: -0.0000

Data Index: 1614392 (The Pile)

Max Activating Token Index: 302

Text #6

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.1749. Min Act: -0.0000

Data Index: 1247756 (The Pile)

Max Activating Token Index: 295

Text #7

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.8921. Min Act: -0.0000

Data Index: 473446 (The Pile)

Max Activating Token Index: 35

Text #8

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.2007. Min Act: -0.0000

Data Index: 1149142 (The Pile)

Max Activating Token Index: 40

Text #9

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.0319. Min Act: -0.0000

Data Index: 98703 (The Pile)

Max Activating Token Index: 108

Text #10

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.1200. Min Act: -0.0000

Data Index: 580586 (The Pile)

Max Activating Token Index: 300

Text #11

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.9796. Min Act: -0.0000

Data Index: 1492999 (The Pile)

Max Activating Token Index: 738

Text #12

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.1049. Min Act: -0.0000

Data Index: 1117338 (The Pile)

Max Activating Token Index: 55

Text #13

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.7916. Min Act: -0.0000

Data Index: 1239275 (The Pile)

Max Activating Token Index: 52

Text #14

Max Range: 5.1797. Min Range: -5.1797

Max Act: 4.1188. Min Act: -0.0000

Data Index: 979700 (The Pile)

Max Activating Token Index: 753

Text #15

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.8807. Min Act: -0.0000

Data Index: 1153706 (The Pile)

Max Activating Token Index: 162

Text #16

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.6938. Min Act: -0.0000

Data Index: 903 (The Pile)

Max Activating Token Index: 557

Text #17

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.9974. Min Act: -0.0000

Data Index: 1326500 (The Pile)

Max Activating Token Index: 245

Text #18

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.6235. Min Act: -0.0000

Data Index: 1705083 (The Pile)

Max Activating Token Index: 187

Text #19

Max Range: 5.1797. Min Range: -5.1797

Max Act: 3.9480. Min Act: -0.0000

Data Index: 1663360 (The Pile)

Max Activating Token Index: 700