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

Dataset: The Pile

Neuron 2920 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.3605. Min Act: -0.0000

Data Index: 1036168 (The Pile)

Max Activating Token Index: 35

Text #1

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.4612. Min Act: -0.0000

Data Index: 1610710 (The Pile)

Max Activating Token Index: 16

Text #2

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.1939. Min Act: -0.0000

Data Index: 1485540 (The Pile)

Max Activating Token Index: 55

Text #3

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.1402. Min Act: -0.0000

Data Index: 700782 (The Pile)

Max Activating Token Index: 67

Text #4

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.1416. Min Act: -0.0000

Data Index: 316861 (The Pile)

Max Activating Token Index: 835

Text #5

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.4033. Min Act: -0.0000

Data Index: 429740 (The Pile)

Max Activating Token Index: 110

Text #6

Max Range: 5.4612. Min Range: -5.4612

Max Act: 5.0014. Min Act: -0.0000

Data Index: 1525695 (The Pile)

Max Activating Token Index: 12

Text #7

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.8416. Min Act: -0.0000

Data Index: 332809 (The Pile)

Max Activating Token Index: 202

Text #8

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.4462. Min Act: -0.0000

Data Index: 1947455 (The Pile)

Max Activating Token Index: 503

Text #9

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.6154. Min Act: -0.0000

Data Index: 178777 (The Pile)

Max Activating Token Index: 15

Text #10

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.7750. Min Act: -0.0000

Data Index: 1153863 (The Pile)

Max Activating Token Index: 237

Text #11

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.7286. Min Act: -0.0000

Data Index: 957966 (The Pile)

Max Activating Token Index: 114

Text #12

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.3434. Min Act: -0.0000

Data Index: 781322 (The Pile)

Max Activating Token Index: 706

Text #13

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.5294. Min Act: -0.0000

Data Index: 1770851 (The Pile)

Max Activating Token Index: 41

Text #14

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.6797. Min Act: -0.0000

Data Index: 1961115 (The Pile)

Max Activating Token Index: 140

Text #15

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.6661. Min Act: -0.0000

Data Index: 1063742 (The Pile)

Max Activating Token Index: 481

Text #16

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.3163. Min Act: -0.0000

Data Index: 1323912 (The Pile)

Max Activating Token Index: 34

Text #17

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.3101. Min Act: -0.0000

Data Index: 753061 (The Pile)

Max Activating Token Index: 19

Text #18

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.6156. Min Act: -0.0000

Data Index: 392902 (The Pile)

Max Activating Token Index: 11

Text #19

Max Range: 5.4612. Min Range: -5.4612

Max Act: 4.4519. Min Act: -0.0000

Data Index: 51978 (The Pile)

Max Activating Token Index: 879